Publications
At ELSA, we aim to inspire and share knowledge within our network and beyond. The collection of publications below provides an overview of both the network’s own output and research we support. Labels distinguish the categories.
Please note that this list makes no claims of being complete. If you have published a paper which is related to ELSA and should be listed, please reach out to our Press and Communications team.
2024
Rai, Shyam Nandan; Cermelli, Fabio; Caputo, Barbara; Masone, Carlo
Mask2Anomaly: Mask Transformer for Universal Open-Set Segmentation Journal Article
In: IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 46, no. 12, pp. 9286-9302, 2024.
@article{10574844,
title = {Mask2Anomaly: Mask Transformer for Universal Open-Set Segmentation},
author = {Shyam Nandan Rai and Fabio Cermelli and Barbara Caputo and Carlo Masone},
doi = {10.1109/TPAMI.2024.3419055},
year = {2024},
date = {2024-01-01},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
volume = {46},
number = {12},
pages = {9286-9302},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Strobel, Volker; Dorigo, Marco; Fritz, Mario
LLM2Swarm: Robot Swarms that Responsively Reason, Plan, and Collaborate through LLMs Proceedings Article
In: NeurIPS 2024 Workshop on Open-World Agents (OWA-2024), 2024.
@inproceedings{StrDorFri2024:neuripsworkshop,
title = {LLM2Swarm: Robot Swarms that Responsively Reason, Plan, and Collaborate through LLMs},
author = {Volker Strobel and Marco Dorigo and Mario Fritz},
url = {https://arxiv.org/abs/2410.11387},
year = {2024},
date = {2024-01-01},
booktitle = {NeurIPS 2024 Workshop on Open-World Agents (OWA-2024)},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Trizna, Dmitrijs; Demetrio, Luca; Biggio, Battista; Roli, Fabio
Living-off-The-Land Reverse-Shell Detection by Informed Data Augmentation Journal Article
In: CoRR, vol. abs/2402.18329, 2024.
@article{DBLP:journals/corr/abs-2402-18329,
title = {Living-off-The-Land Reverse-Shell Detection by Informed Data Augmentation},
author = {Dmitrijs Trizna and Luca Demetrio and Battista Biggio and Fabio Roli},
url = {https://doi.org/10.48550/arXiv.2402.18329},
doi = {10.48550/ARXIV.2402.18329},
year = {2024},
date = {2024-01-01},
journal = {CoRR},
volume = {abs/2402.18329},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Zarlenga, Mateo Espinosa; Collins, Katherine M.; Dvijotham, Krishnamurthy; Weller, Adrian; Shams, Zohreh; Jamnik, Mateja
Learning to Receive Help: Intervention-Aware Concept Embedding Models Miscellaneous
2024.
@misc{zarlenga2024learningreceivehelpinterventionaware,
title = {Learning to Receive Help: Intervention-Aware Concept Embedding Models},
author = {Mateo Espinosa Zarlenga and Katherine M. Collins and Krishnamurthy Dvijotham and Adrian Weller and Zohreh Shams and Mateja Jamnik},
url = {https://arxiv.org/abs/2309.16928},
year = {2024},
date = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Hariat, Marwane; Laurent, Olivier; Kazmierczak, Rémi; Zhang, Shihao; Bursuc, Andrei; Yao, Angela; Franchi, Gianni
Learning to generate training datasets for robust semantic segmentation Proceedings Article
In: 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pp. 3882-3893, 2024, ISSN: 2642-9381.
@inproceedings{10484317,
title = {Learning to generate training datasets for robust semantic segmentation},
author = {Marwane Hariat and Olivier Laurent and Rémi Kazmierczak and Shihao Zhang and Andrei Bursuc and Angela Yao and Gianni Franchi},
doi = {10.1109/WACV57701.2024.00385},
issn = {2642-9381},
year = {2024},
date = {2024-01-01},
booktitle = {2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
pages = {3882-3893},
abstract = {Semantic segmentation methods have advanced significantly. Still, their robustness to real-world perturbations and object types not seen during training remains a challenge, particularly in safety-critical applications. We propose a novel approach to improve the robustness of semantic segmentation techniques by leveraging the synergy between label-to-image generators and image-to-label segmentation models. Specifically, we design Robusta, a novel robust conditional generative adversarial network to generate realistic and plausible perturbed images that can be used to train reliable segmentation models. We conduct in-depth studies of the proposed generative model, assess the performance and robustness of the downstream segmentation network, and demonstrate that our approach can significantly enhance the robustness in the face of real-world perturbations, distribution shifts, and out-of-distribution samples. Our results suggest that this approach could be valuable in safety-critical applications, where the reliability of perception modules such as semantic segmentation is of utmost importance and comes with a limited computational budget in inference. We release our code at github.com/ENSTA-U2IS/robusta.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Semantic segmentation methods have advanced significantly. Still, their robustness to real-world perturbations and object types not seen during training remains a challenge, particularly in safety-critical applications. We propose a novel approach to improve the robustness of semantic segmentation techniques by leveraging the synergy between label-to-image generators and image-to-label segmentation models. Specifically, we design Robusta, a novel robust conditional generative adversarial network to generate realistic and plausible perturbed images that can be used to train reliable segmentation models. We conduct in-depth studies of the proposed generative model, assess the performance and robustness of the downstream segmentation network, and demonstrate that our approach can significantly enhance the robustness in the face of real-world perturbations, distribution shifts, and out-of-distribution samples. Our results suggest that this approach could be valuable in safety-critical applications, where the reliability of perception modules such as semantic segmentation is of utmost importance and comes with a limited computational budget in inference. We release our code at github.com/ENSTA-U2IS/robusta.
Shao, Daqian; Soleymani, Ashkan; Quinzan, Francesco; Kwiatkowska, Marta
Learning decision policies with instrumental variables through double machine learning Proceedings Article
In: Proceedings of the 41st International Conference on Machine Learning, JMLR.org, Vienna, Austria, 2024.
@inproceedings{10.5555/3692070.3693881,
title = {Learning decision policies with instrumental variables through double machine learning},
author = {Daqian Shao and Ashkan Soleymani and Francesco Quinzan and Marta Kwiatkowska},
year = {2024},
date = {2024-01-01},
booktitle = {Proceedings of the 41st International Conference on Machine Learning},
publisher = {JMLR.org},
address = {Vienna, Austria},
series = {ICML'24},
abstract = {A common issue in learning decision-making policies in data-rich settings is spurious correlations in the offline dataset, which can be caused by hidden confounders. Instrumental variable (IV) regression, which utilises a key unconfounded variable known as the instrument, is a standard technique for learning causal relationships between confounded action, outcome, and context variables. Most recent IV regression algorithms use a two-stage approach, where a deep neural network (DNN) estimator learnt in the first stage is directly plugged into the second stage, in which another DNN is used to estimate the causal effect. Naively plugging the estimator can cause heavy bias in the second stage, especially when regularisation bias is present in the first stage estimator. We propose DML-IV, a non-linear IV regression method that reduces the bias in two-stage IV regressions and effectively learns high-performing policies. We derive a novel learning objective to reduce bias and design the DML-IV algorithm following the double/debiased machine learning (DML) framework. The learnt DML-IV estimator has strong convergence rate and O(N-1/2) suboptimality guarantees that match those when the dataset is unconfounded. DML-IV outperforms state-of-the-art IV regression methods on IV regression benchmarks and learns high-performing policies in the presence of instruments.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A common issue in learning decision-making policies in data-rich settings is spurious correlations in the offline dataset, which can be caused by hidden confounders. Instrumental variable (IV) regression, which utilises a key unconfounded variable known as the instrument, is a standard technique for learning causal relationships between confounded action, outcome, and context variables. Most recent IV regression algorithms use a two-stage approach, where a deep neural network (DNN) estimator learnt in the first stage is directly plugged into the second stage, in which another DNN is used to estimate the causal effect. Naively plugging the estimator can cause heavy bias in the second stage, especially when regularisation bias is present in the first stage estimator. We propose DML-IV, a non-linear IV regression method that reduces the bias in two-stage IV regressions and effectively learns high-performing policies. We derive a novel learning objective to reduce bias and design the DML-IV algorithm following the double/debiased machine learning (DML) framework. The learnt DML-IV estimator has strong convergence rate and O(N-1/2) suboptimality guarantees that match those when the dataset is unconfounded. DML-IV outperforms state-of-the-art IV regression methods on IV regression benchmarks and learns high-performing policies in the presence of instruments.
Kapoor, Sanyam; Gruver, Nate; Roberts, Manley; Collins, Katherine; Pal, Arka; Bhatt, Umang; Weller, Adrian; Dooley, Samuel; Goldblum, Micah; Wilson, Andrew Gordon
Large Language Models Must Be Taught to Know What They Don’t Know Miscellaneous
2024.
@misc{kapoor2024largelanguagemodelstaught,
title = {Large Language Models Must Be Taught to Know What They Don't Know},
author = {Sanyam Kapoor and Nate Gruver and Manley Roberts and Katherine Collins and Arka Pal and Umang Bhatt and Adrian Weller and Samuel Dooley and Micah Goldblum and Andrew Gordon Wilson},
url = {https://arxiv.org/abs/2406.08391},
year = {2024},
date = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Anonymous,
Is Mamba Capable of In-Context Learning? Proceedings Article
In: AutoML 2024 Methods Track, 2024.
@inproceedings{anonymous2024is,
title = {Is Mamba Capable of In-Context Learning?},
author = {Anonymous},
url = {https://openreview.net/forum?id=rJhOG0P8nr},
year = {2024},
date = {2024-01-01},
booktitle = {AutoML 2024 Methods Track},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Donghi, Giovanni; Pasa, Luca; Oneto, Luca; Gallicchio, Claudio; Micheli, Alessio; Anguita, Davide; Sperduti, Alessandro; Navarin, Nicolò
Investigating over-parameterized randomized graph networks Journal Article
In: Neurocomputing, vol. 606, pp. 128281, 2024, ISSN: 0925-2312.
@article{DONGHI2024128281,
title = {Investigating over-parameterized randomized graph networks},
author = {Giovanni Donghi and Luca Pasa and Luca Oneto and Claudio Gallicchio and Alessio Micheli and Davide Anguita and Alessandro Sperduti and Nicolò Navarin},
url = {https://www.sciencedirect.com/science/article/pii/S092523122401052X},
doi = {https://doi.org/10.1016/j.neucom.2024.128281},
issn = {0925-2312},
year = {2024},
date = {2024-01-01},
journal = {Neurocomputing},
volume = {606},
pages = {128281},
abstract = {In this paper, we investigate neural models based on graph random features for classification tasks. First, we aim to understand when over parameterization, namely generating more features than the ones necessary to interpolate, may be beneficial for the generalization abilities of the resulting models. We employ two measures: one from the algorithmic stability framework and another one based on information theory. We provide empirical evidence from several commonly adopted graph datasets showing that the considered measures, even without considering task labels, can be effective for this purpose. Additionally, we investigate whether these measures can aid in the process of hyperparameters selection. The results of our empirical analysis show that the considered measures have good correlations with the estimated generalization performance of the models with different hyperparameter configurations. Moreover, they can be used to identify good hyperparameters, achieving results comparable to the ones obtained with a classic grid search.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this paper, we investigate neural models based on graph random features for classification tasks. First, we aim to understand when over parameterization, namely generating more features than the ones necessary to interpolate, may be beneficial for the generalization abilities of the resulting models. We employ two measures: one from the algorithmic stability framework and another one based on information theory. We provide empirical evidence from several commonly adopted graph datasets showing that the considered measures, even without considering task labels, can be effective for this purpose. Additionally, we investigate whether these measures can aid in the process of hyperparameters selection. The results of our empirical analysis show that the considered measures have good correlations with the estimated generalization performance of the models with different hyperparameter configurations. Moreover, they can be used to identify good hyperparameters, achieving results comparable to the ones obtained with a classic grid search.
Wallace, Benedikte; Otterdijk, Marieke; Zhang, Yuchong; Rajabi, Nona; Marin-Bucio, Diego; Kragic, Danica; Torresen, Jim
Imitation or Innovation? Translating Features of Expressive Motion from Humans to Robots Proceedings Article
In: Proceedings of the 12th International Conference on Human-Agent Interaction, pp. 296–304, Association for Computing Machinery, Swansea, United Kingdom, 2024, ISBN: 9798400711787.
@inproceedings{10.1145/3687272.3688302,
title = {Imitation or Innovation? Translating Features of Expressive Motion from Humans to Robots},
author = {Benedikte Wallace and Marieke Otterdijk and Yuchong Zhang and Nona Rajabi and Diego Marin-Bucio and Danica Kragic and Jim Torresen},
url = {https://doi.org/10.1145/3687272.3688302},
doi = {10.1145/3687272.3688302},
isbn = {9798400711787},
year = {2024},
date = {2024-01-01},
booktitle = {Proceedings of the 12th International Conference on Human-Agent Interaction},
pages = {296–304},
publisher = {Association for Computing Machinery},
address = {Swansea, United Kingdom},
series = {HAI '24},
abstract = {Expressive robot motion can help establish acceptance of this technology in everyday life, but understanding what makes movement expressive is a complex and multifaceted task. This paper presents the results of an online study with 46 participants, it aims to explore how people perceive and interpret the expressive qualities of human movement and how they envision the translation of their description into an imagined non-humanoid, quadrupedal robot. Through a qualitative analysis of responses, we conceptualize three themes: their understanding of intent, their interpretations of movement qualities, and finally, their translation from human to robot movement. Respondents’ descriptions of their initial understanding of the performer’s intent fall into two modes, bio-mechanical and narrative. We illustrate their interpretations of movement qualities through four strategies: movement features as kinematic indicators, intent indicators, attributed context, and perceived internal states. Lastly, we observe their translation from human to robot movement, with a particular focus on respondents’ use of kinaesthetic empathy and anthropomorphism. Our findings aim to support a bottom-up approach, using users’ general knowledge for designing expressive robot motion.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Expressive robot motion can help establish acceptance of this technology in everyday life, but understanding what makes movement expressive is a complex and multifaceted task. This paper presents the results of an online study with 46 participants, it aims to explore how people perceive and interpret the expressive qualities of human movement and how they envision the translation of their description into an imagined non-humanoid, quadrupedal robot. Through a qualitative analysis of responses, we conceptualize three themes: their understanding of intent, their interpretations of movement qualities, and finally, their translation from human to robot movement. Respondents’ descriptions of their initial understanding of the performer’s intent fall into two modes, bio-mechanical and narrative. We illustrate their interpretations of movement qualities through four strategies: movement features as kinematic indicators, intent indicators, attributed context, and perceived internal states. Lastly, we observe their translation from human to robot movement, with a particular focus on respondents’ use of kinaesthetic empathy and anthropomorphism. Our findings aim to support a bottom-up approach, using users’ general knowledge for designing expressive robot motion.
Zhang, Ziyang; Angelov, Plamen P.; Kangin, Dmitry; Long’ep’e, Nicolas
IMAFD: An Interpretable Multi-stage Approach to Flood Detection from time series Multispectral Data Journal Article
In: ArXiv, vol. abs/2405.07916, 2024.
@article{Zhang2024IMAFDAI,
title = {IMAFD: An Interpretable Multi-stage Approach to Flood Detection from time series Multispectral Data},
author = {Ziyang Zhang and Plamen P. Angelov and Dmitry Kangin and Nicolas Long'ep'e},
url = {https://api.semanticscholar.org/CorpusID:269757928},
year = {2024},
date = {2024-01-01},
journal = {ArXiv},
volume = {abs/2405.07916},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Sheth, Ivaxi; Abdelnabi, Sahar; Fritz, Mario
Hypothesizing Missing Causal Variables with LLMs Miscellaneous
2024.
@misc{sheth2024hypothesizingmissingcausalvariables,
title = {Hypothesizing Missing Causal Variables with LLMs},
author = {Ivaxi Sheth and Sahar Abdelnabi and Mario Fritz},
url = {https://arxiv.org/abs/2409.02604},
year = {2024},
date = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Zhang, Yuchong; Kassem, Khaled; Gong, Zhengya; Mo, Fan; Ma, Yong; Kirjavainen, Emma; Häkkilä, Jonna
Human-centered AI Technologies in Human-robot Interaction for Social Settings Proceedings Article
In: Proceedings of the International Conference on Mobile and Ubiquitous Multimedia, pp. 501–505, Association for Computing Machinery, New York, NY, USA, 2024, ISBN: 9798400712838.
@inproceedings{10.1145/3701571.3701610,
title = {Human-centered AI Technologies in Human-robot Interaction for Social Settings},
author = {Yuchong Zhang and Khaled Kassem and Zhengya Gong and Fan Mo and Yong Ma and Emma Kirjavainen and Jonna Häkkilä},
url = {https://doi.org/10.1145/3701571.3701610},
doi = {10.1145/3701571.3701610},
isbn = {9798400712838},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Proceedings of the International Conference on Mobile and Ubiquitous Multimedia},
pages = {501–505},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
series = {MUM '24},
abstract = {The increasing integration of human-robot interaction (HRI) into social settings demands the development of human-centered AI technologies that prioritize intuitive, ethical, and empathetic interactions. As robots become more prevalent in everyday life—ranging from assistive devices in healthcare to educational tools in classrooms and customer service agents in retail—it is essential to ensure they can communicate and collaborate with humans in ways that are not only effective but also socially appropriate and meaningful. This workshop aims to explore cutting-edge advancements and interdisciplinary approaches to building AI-driven systems that facilitate effective, meaningful, and socially appropriate interactions between robots and humans across various environments such as healthcare, education, and customer service. We will primarily focus on several key themes, such as human-centered contextual AI, AI-driven intelligent robotics, ethical and responsible AI, and real-world applications. This workshop invites contributions from researchers, practitioners, and developers who are working on AI systems that empower robots to operate effectively in human-centered environments. By addressing challenges such as interpreting human emotions, understanding social cues, and adhering to ethical standards, and by sharing advancements in human-centered AI, we aim to shape the future of HRI. Our goal is to ensure that robots enrich human social experiences, fostering interactions that are not only efficient but also enhance the quality of life. By uniting efforts from various disciplines, we aspire to create robots that seamlessly integrate into society, ultimately contributing to a more harmonious coexistence between humans and robotic systems.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The increasing integration of human-robot interaction (HRI) into social settings demands the development of human-centered AI technologies that prioritize intuitive, ethical, and empathetic interactions. As robots become more prevalent in everyday life—ranging from assistive devices in healthcare to educational tools in classrooms and customer service agents in retail—it is essential to ensure they can communicate and collaborate with humans in ways that are not only effective but also socially appropriate and meaningful. This workshop aims to explore cutting-edge advancements and interdisciplinary approaches to building AI-driven systems that facilitate effective, meaningful, and socially appropriate interactions between robots and humans across various environments such as healthcare, education, and customer service. We will primarily focus on several key themes, such as human-centered contextual AI, AI-driven intelligent robotics, ethical and responsible AI, and real-world applications. This workshop invites contributions from researchers, practitioners, and developers who are working on AI systems that empower robots to operate effectively in human-centered environments. By addressing challenges such as interpreting human emotions, understanding social cues, and adhering to ethical standards, and by sharing advancements in human-centered AI, we aim to shape the future of HRI. Our goal is to ensure that robots enrich human social experiences, fostering interactions that are not only efficient but also enhance the quality of life. By uniting efforts from various disciplines, we aspire to create robots that seamlessly integrate into society, ultimately contributing to a more harmonious coexistence between humans and robotic systems.
Lorenz, T; Kwiatkowska, M; Fritz, M
FullCert: deterministic end-to-end certification for training and inference of neural networks Proceedings Article
In: Springer, 2024.
@inproceedings{lorenz2024a,
title = {FullCert: deterministic end-to-end certification for training and inference of neural networks},
author = {T Lorenz and M Kwiatkowska and M Fritz},
year = {2024},
date = {2024-01-01},
publisher = {Springer},
series = {Lecture Notes in Computer Science},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Borlino, Francesco Cappio; Lu, Lorenzo; Tommasi, Tatiana
Foundation Models and Fine-Tuning: A Benchmark for Out of Distribution Detection Journal Article
In: IEEE Access, vol. 12, pp. 79401-79414, 2024, ISSN: 2169-3536.
@article{10547247,
title = {Foundation Models and Fine-Tuning: A Benchmark for Out of Distribution Detection},
author = {Francesco Cappio Borlino and Lorenzo Lu and Tatiana Tommasi},
doi = {10.1109/ACCESS.2024.3409587},
issn = {2169-3536},
year = {2024},
date = {2024-01-01},
journal = {IEEE Access},
volume = {12},
pages = {79401-79414},
abstract = {The rise of foundation models is pushing Computer Vision research towards a paradigm shift, in the wake of what already happened in the Natural Language Processing field. These models, trained at scale on huge data collections, provide high-quality representations that generalize well enough to be applied directly to downstream tasks, often outperforming task-specific models. The Out Of Distribution (OOD) detection problem, which involves the ability to recognize when test samples come from a previously unseen semantic category, represents one of the research fields in which this paradigm shift could have the greatest impact. However, existing testbeds are limited in scale and scope and get easily saturated when adopting foundation-based pretrainings. With this work, we introduce a new benchmark covering realistic yet harder OOD detection tasks to properly assess the performance of large pretrained models. We design an experimental framework to analyze specific choices in the model learning and use (which dataset, pretraining objective, OOD scoring function) and extensively evaluate the comparison to standard approaches that leverage a training phase on the available In Distribution (ID) data. The results highlight the actual performance benefits of leveraging foundation models in this context without any further learning effort, and identify situations where task-specific fine-tuning remains the best choice.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The rise of foundation models is pushing Computer Vision research towards a paradigm shift, in the wake of what already happened in the Natural Language Processing field. These models, trained at scale on huge data collections, provide high-quality representations that generalize well enough to be applied directly to downstream tasks, often outperforming task-specific models. The Out Of Distribution (OOD) detection problem, which involves the ability to recognize when test samples come from a previously unseen semantic category, represents one of the research fields in which this paradigm shift could have the greatest impact. However, existing testbeds are limited in scale and scope and get easily saturated when adopting foundation-based pretrainings. With this work, we introduce a new benchmark covering realistic yet harder OOD detection tasks to properly assess the performance of large pretrained models. We design an experimental framework to analyze specific choices in the model learning and use (which dataset, pretraining objective, OOD scoring function) and extensively evaluate the comparison to standard approaches that leverage a training phase on the available In Distribution (ID) data. The results highlight the actual performance benefits of leveraging foundation models in this context without any further learning effort, and identify situations where task-specific fine-tuning remains the best choice.
Iurada, Leonardo; Ciccone, Marco; Tommasi, Tatiana
Finding Lottery Tickets in Vision Models via Data-Driven Spectral Foresight Pruning Journal Article
In: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 16142-16151, 2024.
@article{Iurada2024FindingLT,
title = {Finding Lottery Tickets in Vision Models via Data-Driven Spectral Foresight Pruning},
author = {Leonardo Iurada and Marco Ciccone and Tatiana Tommasi},
url = {https://api.semanticscholar.org/CorpusID:270226621},
year = {2024},
date = {2024-01-01},
journal = {2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
pages = {16142-16151},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Wang, Hui-Po; Chen, Dingfan; Kerkouche, Raouf; Fritz, Mario
FedLAP-DP: Federated Learning by Sharing Differentially Private Loss Approximations Miscellaneous
2024.
@misc{wang2024fedlapdpfederatedlearningsharing,
title = {FedLAP-DP: Federated Learning by Sharing Differentially Private Loss Approximations},
author = {Hui-Po Wang and Dingfan Chen and Raouf Kerkouche and Mario Fritz},
url = {https://arxiv.org/abs/2302.01068},
year = {2024},
date = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Chen, Jialuo; Wang, Jingyi; Zhang, Xiyue; Sun, Youcheng; Kwiatkowska, Marta; Chen, Jiming; Cheng, Peng
FAST: Boosting Uncertainty-based Test Prioritization Methods for Neural Networks via Feature Selection Proceedings Article
In: Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering, pp. 895–906, Association for Computing Machinery, Sacramento, CA, USA, 2024, ISBN: 9798400712487.
@inproceedings{10.1145/3691620.3695472,
title = {FAST: Boosting Uncertainty-based Test Prioritization Methods for Neural Networks via Feature Selection},
author = {Jialuo Chen and Jingyi Wang and Xiyue Zhang and Youcheng Sun and Marta Kwiatkowska and Jiming Chen and Peng Cheng},
url = {https://doi.org/10.1145/3691620.3695472},
doi = {10.1145/3691620.3695472},
isbn = {9798400712487},
year = {2024},
date = {2024-01-01},
booktitle = {Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering},
pages = {895–906},
publisher = {Association for Computing Machinery},
address = {Sacramento, CA, USA},
series = {ASE '24},
abstract = {Due to the vast testing space, the increasing demand for effective and efficient testing of deep neural networks (DNNs) has led to the development of various DNN test case prioritization techniques. However, the fact that DNNs can deliver high-confidence predictions for incorrectly predicted examples, known as the over-confidence problem, causes these methods to fail to reveal high-confidence errors. To address this limitation, in this work, we propose FAST, a method that boosts existing prioritization methods through guided FeAture SelecTion. FAST is based on the insight that certain features may introduce noise that affects the model's output confidence, thereby contributing to high-confidence errors. It quantifies the importance of each feature for the model's correct predictions, and then dynamically prunes the information from the noisy features during inference to derive a new probability vector for the uncertainty estimation. With the help of FAST, the high-confidence errors and correctly classified examples become more distinguishable, resulting in higher APFD (Average Percentage of Fault Detection) values for test prioritization, and higher generalization ability for model enhancement. We conduct extensive experiments to evaluate FAST across a diverse set of model structures on multiple benchmark datasets to validate the effectiveness, efficiency, and scalability of FAST compared to the state-of-the-art prioritization techniques.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Due to the vast testing space, the increasing demand for effective and efficient testing of deep neural networks (DNNs) has led to the development of various DNN test case prioritization techniques. However, the fact that DNNs can deliver high-confidence predictions for incorrectly predicted examples, known as the over-confidence problem, causes these methods to fail to reveal high-confidence errors. To address this limitation, in this work, we propose FAST, a method that boosts existing prioritization methods through guided FeAture SelecTion. FAST is based on the insight that certain features may introduce noise that affects the model’s output confidence, thereby contributing to high-confidence errors. It quantifies the importance of each feature for the model’s correct predictions, and then dynamically prunes the information from the noisy features during inference to derive a new probability vector for the uncertainty estimation. With the help of FAST, the high-confidence errors and correctly classified examples become more distinguishable, resulting in higher APFD (Average Percentage of Fault Detection) values for test prioritization, and higher generalization ability for model enhancement. We conduct extensive experiments to evaluate FAST across a diverse set of model structures on multiple benchmark datasets to validate the effectiveness, efficiency, and scalability of FAST compared to the state-of-the-art prioritization techniques.
Iurada, Leonardo; Bucci, Silvia; Hospedales, Timothy M.; Tommasi, Tatiana
Fairness Meets Cross-Domain Learning: A Benchmark of Models and Metrics Journal Article
In: IEEE Access, vol. 12, pp. 47854-47867, 2024, ISSN: 2169-3536.
@article{10487929,
title = {Fairness Meets Cross-Domain Learning: A Benchmark of Models and Metrics},
author = {Leonardo Iurada and Silvia Bucci and Timothy M. Hospedales and Tatiana Tommasi},
doi = {10.1109/ACCESS.2024.3383841},
issn = {2169-3536},
year = {2024},
date = {2024-01-01},
journal = {IEEE Access},
volume = {12},
pages = {47854-47867},
abstract = {Deep learning-based recognition systems are deployed at scale for real-world applications that inevitably involve our social life. Although of great support when making complex decisions, they might capture spurious data correlations and leverage sensitive attributes (e.g., age, gender, ethnicity). How to factor out this information while maintaining high performance is a problem with several open questions, many of which are shared with those of the domain adaptation and generalization literature which aims at avoiding visual domain biases. In this work, we propose an in-depth study of the relationship between cross-domain learning (CD) and model fairness, by experimentally evaluating 14 CD approaches together with 3 state-of-the-art fairness algorithms on 5 datasets of faces and medical images spanning several demographic groups. We consider attribute classification and landmark detection tasks: the latter is introduced here for the first time in the fairness literature, showing how keypoint localization may be affected by sensitive attribute biases. To assess the analyzed methods, we adopt widely used evaluation metrics while also presenting their limits with a detailed review. Moreover, we propose a new Harmonic Fairness (HF) score that can ease unfairness mitigation model comparisons. Overall, our work shows how CD approaches can outperform state-of-the-art fairness algorithms and defines a framework with dataset and metrics as well as a code suite to pave the way for a more systematic analysis of fairness problems in computer vision (Code available at: https://github.com/iurada/fairness_crossdomain).},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Deep learning-based recognition systems are deployed at scale for real-world applications that inevitably involve our social life. Although of great support when making complex decisions, they might capture spurious data correlations and leverage sensitive attributes (e.g., age, gender, ethnicity). How to factor out this information while maintaining high performance is a problem with several open questions, many of which are shared with those of the domain adaptation and generalization literature which aims at avoiding visual domain biases. In this work, we propose an in-depth study of the relationship between cross-domain learning (CD) and model fairness, by experimentally evaluating 14 CD approaches together with 3 state-of-the-art fairness algorithms on 5 datasets of faces and medical images spanning several demographic groups. We consider attribute classification and landmark detection tasks: the latter is introduced here for the first time in the fairness literature, showing how keypoint localization may be affected by sensitive attribute biases. To assess the analyzed methods, we adopt widely used evaluation metrics while also presenting their limits with a detailed review. Moreover, we propose a new Harmonic Fairness (HF) score that can ease unfairness mitigation model comparisons. Overall, our work shows how CD approaches can outperform state-of-the-art fairness algorithms and defines a framework with dataset and metrics as well as a code suite to pave the way for a more systematic analysis of fairness problems in computer vision (Code available at: https://github.com/iurada/fairness_crossdomain).
Franco, Danilo; D’Amato, Vincenzo Stefano; Pasa, Luca; Navarin, Nicolò; Oneto, Luca
Fair graph representation learning: Empowering NIFTY via Biased Edge Dropout and Fair Attribute Preprocessing Journal Article
In: Neurocomputing, vol. 563, pp. 126948, 2024, ISSN: 0925-2312.
@article{FRANCO2024126948,
title = {Fair graph representation learning: Empowering NIFTY via Biased Edge Dropout and Fair Attribute Preprocessing},
author = {Danilo Franco and Vincenzo Stefano D’Amato and Luca Pasa and Nicolò Navarin and Luca Oneto},
url = {https://www.sciencedirect.com/science/article/pii/S0925231223010718},
doi = {https://doi.org/10.1016/j.neucom.2023.126948},
issn = {0925-2312},
year = {2024},
date = {2024-01-01},
journal = {Neurocomputing},
volume = {563},
pages = {126948},
abstract = {The increasing complexity and amount of data available in modern applications strongly demand Trustworthy Learning algorithms that can be fed directly with complex and large graphs data. In fact, on one hand, machine learning models must meet high technical standards (e.g., high accuracy with limited computational requirements), but, at the same time, they must be sure not to discriminate against subgroups of the population (e.g., based on gender or ethnicity). Graph Neural Networks (GNNs) are currently the most effective solution to meet the technical requirements, even if it has been demonstrated that they inherit and amplify the biases contained in the data as a reflection of societal inequities. In fact, when dealing with graph data, these biases can be hidden not only in the node attributes but also in the connections between entities. Several Fair GNNs have been proposed in the literature, with uNIfying Fairness and stabiliTY (NIFTY) (Agarwal et al., 2021) being one of the most effective. In this paper, we will empower NIFTY’s fairness with two new strategies. The first one is a Biased Edge Dropout, namely, we drop graph edges to balance homophilous and heterophilous sensitive connections, mitigating the bias induced by subgroup node cardinality. The second one is Attributes Preprocessing, which is the process of learning a fair transformation of the original node attributes. The effectiveness of our proposal will be tested on a series of datasets with increasingly challenging scenarios. These scenarios will deal with different levels of knowledge about the entire graph, i.e., how many portions of the graph are known and which sub-portion is labelled at the training and forward phases.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The increasing complexity and amount of data available in modern applications strongly demand Trustworthy Learning algorithms that can be fed directly with complex and large graphs data. In fact, on one hand, machine learning models must meet high technical standards (e.g., high accuracy with limited computational requirements), but, at the same time, they must be sure not to discriminate against subgroups of the population (e.g., based on gender or ethnicity). Graph Neural Networks (GNNs) are currently the most effective solution to meet the technical requirements, even if it has been demonstrated that they inherit and amplify the biases contained in the data as a reflection of societal inequities. In fact, when dealing with graph data, these biases can be hidden not only in the node attributes but also in the connections between entities. Several Fair GNNs have been proposed in the literature, with uNIfying Fairness and stabiliTY (NIFTY) (Agarwal et al., 2021) being one of the most effective. In this paper, we will empower NIFTY’s fairness with two new strategies. The first one is a Biased Edge Dropout, namely, we drop graph edges to balance homophilous and heterophilous sensitive connections, mitigating the bias induced by subgroup node cardinality. The second one is Attributes Preprocessing, which is the process of learning a fair transformation of the original node attributes. The effectiveness of our proposal will be tested on a series of datasets with increasingly challenging scenarios. These scenarios will deal with different levels of knowledge about the entire graph, i.e., how many portions of the graph are known and which sub-portion is labelled at the training and forward phases.
Baader, Maximilian; Mueller, Mark Niklas; Mao, Yuhao; Vechev, Martin
Expressivity of ReLU-Networks under Convex Relaxations Proceedings Article
In: The Twelfth International Conference on Learning Representations, 2024.
@inproceedings{baader2024expressivity,
title = {Expressivity of ReLU-Networks under Convex Relaxations},
author = {Maximilian Baader and Mark Niklas Mueller and Yuhao Mao and Martin Vechev},
url = {https://openreview.net/forum?id=awHTL3Hpto},
year = {2024},
date = {2024-01-01},
booktitle = {The Twelfth International Conference on Learning Representations},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Li, Yi; Angelov, Plamen P
Explainable Audio-Visual Representation Learning via Prototypical Contrastive Masked Autoencoder Proceedings Article
In: NeurIPS 2024 Workshop: Self-Supervised Learning – Theory and Practice, 2024.
@inproceedings{li2024explainable,
title = {Explainable Audio-Visual Representation Learning via Prototypical Contrastive Masked Autoencoder},
author = {Yi Li and Plamen P Angelov},
url = {https://openreview.net/forum?id=ZLPVS0T4lc},
year = {2024},
date = {2024-01-01},
booktitle = {NeurIPS 2024 Workshop: Self-Supervised Learning - Theory and Practice},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Berton, Gabriele; Stoken, Alex; Caputo, Barbara; Masone, Carlo
EarthLoc: Astronaut Photography Localization by Indexing Earth from Space Proceedings Article
In: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 12754-12764, 2024.
@inproceedings{10656123,
title = {EarthLoc: Astronaut Photography Localization by Indexing Earth from Space},
author = {Gabriele Berton and Alex Stoken and Barbara Caputo and Carlo Masone},
doi = {10.1109/CVPR52733.2024.01212},
year = {2024},
date = {2024-01-01},
booktitle = {2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
pages = {12754-12764},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Fan, Simin; Pagliardini, Matteo; Jaggi, Martin
DOGE: domain reweighting with generalization estimation Proceedings Article
In: Proceedings of the 41st International Conference on Machine Learning, JMLR.org, Vienna, Austria, 2024.
@inproceedings{10.5555/3692070.3692585,
title = {DOGE: domain reweighting with generalization estimation},
author = {Simin Fan and Matteo Pagliardini and Martin Jaggi},
year = {2024},
date = {2024-01-01},
booktitle = {Proceedings of the 41st International Conference on Machine Learning},
publisher = {JMLR.org},
address = {Vienna, Austria},
series = {ICML'24},
abstract = {The coverage and composition of the pretraining data significantly impacts the generalization ability of Large Language Models (LLMs). Despite its importance, recent LLMs still rely on heuristics and trial and error to increase or reduce the influence of data-domains. We propose DOmain reweighting with Generalization Estimation (DOGE), which optimizes the probability of sampling from each domain (domain weights) in a principled way. Our approach is a two-stage process consisting of (i) training a proxy model to obtain domain weights using a bi-level optimization algorithm; (ii) training a larger base model by sampling training domains according to the learned domain weights. In our experiments, we extensively show how DOGE improves the generalization of the base model to any target data mixture. On the SlimPajama dataset, our base model gets better perplexity and few-shot reasoning accuracies across 6 tasks compared to baseline methods. Moreover, aiming to generalize to out-of-domain target tasks, which is unseen in the pretraining corpus (OOD domain), DOGE can effectively identify inter-domain dependencies, and consistently achieves better test perplexity on the target domain. We provide the codebase at https://github.com/Olivia-fsm/doge.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The coverage and composition of the pretraining data significantly impacts the generalization ability of Large Language Models (LLMs). Despite its importance, recent LLMs still rely on heuristics and trial and error to increase or reduce the influence of data-domains. We propose DOmain reweighting with Generalization Estimation (DOGE), which optimizes the probability of sampling from each domain (domain weights) in a principled way. Our approach is a two-stage process consisting of (i) training a proxy model to obtain domain weights using a bi-level optimization algorithm; (ii) training a larger base model by sampling training domains according to the learned domain weights. In our experiments, we extensively show how DOGE improves the generalization of the base model to any target data mixture. On the SlimPajama dataset, our base model gets better perplexity and few-shot reasoning accuracies across 6 tasks compared to baseline methods. Moreover, aiming to generalize to out-of-domain target tasks, which is unseen in the pretraining corpus (OOD domain), DOGE can effectively identify inter-domain dependencies, and consistently achieves better test perplexity on the target domain. We provide the codebase at https://github.com/Olivia-fsm/doge.
Cesa-Bianchi, Nicolò; Cesari, Tommaso; Vecchia, Riccardo Della
Cooperative Online Learning with Feedback Graphs Journal Article
In: Transactions on Machine Learning Research, 2024, ISSN: 2835-8856.
@article{cesa-bianchi2024cooperative,
title = {Cooperative Online Learning with Feedback Graphs},
author = {Nicolò Cesa-Bianchi and Tommaso Cesari and Riccardo Della Vecchia},
url = {https://openreview.net/forum?id=PtNyIboDIG},
issn = {2835-8856},
year = {2024},
date = {2024-01-01},
journal = {Transactions on Machine Learning Research},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Abdelnabi, Sahar; Gomaa, Amr; Sivaprasad, Sarath; Schönherr, Lea; Fritz, Mario
Cooperation, Competition, and Maliciousness: LLM-Stakeholders Interactive Negotiation Miscellaneous
2024.
@misc{abdelnabi2024cooperationcompetitionmaliciousnessllmstakeholders,
title = {Cooperation, Competition, and Maliciousness: LLM-Stakeholders Interactive Negotiation},
author = {Sahar Abdelnabi and Amr Gomaa and Sarath Sivaprasad and Lea Schönherr and Mario Fritz},
url = {https://arxiv.org/abs/2309.17234},
year = {2024},
date = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Shamsabadi, Ali Shahin; Tan, Gefei; Cebere, Tudor Ioan; Bellet, Aurélien; Haddadi, Hamed; Papernot, Nicolas; Wang, Xiao; Weller, Adrian
Confidential-DPproof: Confidential Proof of Differentially Private Training Proceedings Article
In: The Twelfth International Conference on Learning Representations, 2024.
@inproceedings{shamsabadi2024confidentialdpproof,
title = {Confidential-DPproof: Confidential Proof of Differentially Private Training},
author = {Ali Shahin Shamsabadi and Gefei Tan and Tudor Ioan Cebere and Aurélien Bellet and Hamed Haddadi and Nicolas Papernot and Xiao Wang and Adrian Weller},
url = {https://openreview.net/forum?id=PQY2v6VtGe},
year = {2024},
date = {2024-01-01},
booktitle = {The Twelfth International Conference on Learning Representations},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Li, Yi; Sun, Yang; Angelov, Plamen
Complex-Cycle-Consistent Diffusion Model for Monaural Speech Enhancement Journal Article
In: CoRR, vol. abs/2412.08856, 2024.
@article{DBLP:journals/corr/abs-2412-08856,
title = {Complex-Cycle-Consistent Diffusion Model for Monaural Speech Enhancement},
author = {Yi Li and Yang Sun and Plamen Angelov},
url = {https://doi.org/10.48550/arXiv.2412.08856},
doi = {10.48550/ARXIV.2412.08856},
year = {2024},
date = {2024-01-01},
journal = {CoRR},
volume = {abs/2412.08856},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Kumar, Sreejan; Marjieh, Raja; Zhang, Byron; Campbell, Declan; Hu, Michael Y.; Bhatt, Umang; Lake, Brenden; Griffiths, Thomas L.
Comparing Abstraction in Humans and Large Language Models Using Multimodal Serial Reproduction Miscellaneous
2024.
@misc{kumar2024comparingabstractionhumanslarge,
title = {Comparing Abstraction in Humans and Large Language Models Using Multimodal Serial Reproduction},
author = {Sreejan Kumar and Raja Marjieh and Byron Zhang and Declan Campbell and Michael Y. Hu and Umang Bhatt and Brenden Lake and Thomas L. Griffiths},
url = {https://arxiv.org/abs/2402.03618},
year = {2024},
date = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Hashemi, Diba; He, Lie; Jaggi, Martin
CoBo: Collaborative Learning via Bilevel Optimization Proceedings Article
In: Globerson, A.; Mackey, L.; Belgrave, D.; Fan, A.; Paquet, U.; Tomczak, J.; Zhang, C. (Ed.): Advances in Neural Information Processing Systems, pp. 15550–15574, Curran Associates, Inc., 2024.
@inproceedings{NEURIPS2024_1c32452f,
title = {CoBo: Collaborative Learning via Bilevel Optimization},
author = {Diba Hashemi and Lie He and Martin Jaggi},
editor = {A. Globerson and L. Mackey and D. Belgrave and A. Fan and U. Paquet and J. Tomczak and C. Zhang},
url = {https://proceedings.neurips.cc/paper_files/paper/2024/file/1c32452f112719f7c1db6d983d060f78-Paper-Conference.pdf},
year = {2024},
date = {2024-01-01},
booktitle = {Advances in Neural Information Processing Systems},
volume = {37},
pages = {15550–15574},
publisher = {Curran Associates, Inc.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Zverev, Egor; Abdelnabi, Sahar; Fritz, Mario; Lampert, Christoph H.
Can LLMs Separate Instructions From Data? And What Do We Even Mean By That? Proceedings Article
In: ICLR 2024 Workshop on Secure and Trustworthy Large Language Models, 2024.
@inproceedings{zverev2024can,
title = {Can LLMs Separate Instructions From Data? And What Do We Even Mean By That?},
author = {Egor Zverev and Sahar Abdelnabi and Mario Fritz and Christoph H. Lampert},
url = {https://openreview.net/forum?id=32eytC1Nt1},
year = {2024},
date = {2024-01-01},
booktitle = {ICLR 2024 Workshop on Secure and Trustworthy Large Language Models},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Collins, Katherine M.; Sucholutsky, Ilia; Bhatt, Umang; Chandra, Kartik; Wong, Lionel; Lee, Mina; Zhang, Cedegao E.; Zhi-Xuan, Tan; Ho, Mark; Mansinghka, Vikash; Weller, Adrian; Tenenbaum, Joshua B.; Griffiths, Thomas L.
Building Machines that Learn and Think with People Miscellaneous
2024.
@misc{collins2024buildingmachineslearnthink,
title = {Building Machines that Learn and Think with People},
author = {Katherine M. Collins and Ilia Sucholutsky and Umang Bhatt and Kartik Chandra and Lionel Wong and Mina Lee and Cedegao E. Zhang and Tan Zhi-Xuan and Mark Ho and Vikash Mansinghka and Adrian Weller and Joshua B. Tenenbaum and Thomas L. Griffiths},
url = {https://arxiv.org/abs/2408.03943},
year = {2024},
date = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Wicker, Matthew; Patane, Andrea; Laurenti, Luca; Kwiatkowska, Marta
Adversarial Robustness Certification for Bayesian Neural Networks Proceedings Article
In: Formal Methods: 26th International Symposium, FM 2024, Milan, Italy, September 9–13, 2024, Proceedings, Part I, pp. 3–28, Springer-Verlag, Milan, Italy, 2024, ISBN: 978-3-031-71161-9.
@inproceedings{10.1007/978-3-031-71162-6_1,
title = {Adversarial Robustness Certification for Bayesian Neural Networks},
author = {Matthew Wicker and Andrea Patane and Luca Laurenti and Marta Kwiatkowska},
url = {https://doi.org/10.1007/978-3-031-71162-6_1},
doi = {10.1007/978-3-031-71162-6_1},
isbn = {978-3-031-71161-9},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Formal Methods: 26th International Symposium, FM 2024, Milan, Italy, September 9–13, 2024, Proceedings, Part I},
pages = {3–28},
publisher = {Springer-Verlag},
address = {Milan, Italy},
abstract = {We study the problem of certifying the robustness of Bayesian neural networks (BNNs) to adversarial input perturbations. Specifically, we define two notions of robustness for BNNs in an adversarial setting: probabilistic robustness and decision robustness. The former deals with the probabilistic behaviour of the network, that is, it ensures robustness across different stochastic realisations of the network, while the latter provides guarantees for the overall (output) decision of the BNN. Although these robustness properties cannot be computed analytically, we present a unified computational framework for efficiently and formally bounding them. Our approach is based on weight interval sampling, integration and bound propagation techniques, and can be applied to BNNs with a large number of parameters independently of the (approximate) inference method employed to train the BNN. We evaluate the effectiveness of our method on tasks including airborne collision avoidance, medical imaging and autonomous driving, demonstrating that it can compute non-trivial guarantees on medium size images (i.e., over 16 thousand input parameters).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We study the problem of certifying the robustness of Bayesian neural networks (BNNs) to adversarial input perturbations. Specifically, we define two notions of robustness for BNNs in an adversarial setting: probabilistic robustness and decision robustness. The former deals with the probabilistic behaviour of the network, that is, it ensures robustness across different stochastic realisations of the network, while the latter provides guarantees for the overall (output) decision of the BNN. Although these robustness properties cannot be computed analytically, we present a unified computational framework for efficiently and formally bounding them. Our approach is based on weight interval sampling, integration and bound propagation techniques, and can be applied to BNNs with a large number of parameters independently of the (approximate) inference method employed to train the BNN. We evaluate the effectiveness of our method on tasks including airborne collision avoidance, medical imaging and autonomous driving, demonstrating that it can compute non-trivial guarantees on medium size images (i.e., over 16 thousand input parameters).
Sussex, Scott; Sessa, Pier Giuseppe; Makarova, Anastasia; Krause, Andreas
Adversarial Causal Bayesian Optimization Proceedings Article
In: The Twelfth International Conference on Learning Representations, 2024.
@inproceedings{sussex2024adversarial,
title = {Adversarial Causal Bayesian Optimization},
author = {Scott Sussex and Pier Giuseppe Sessa and Anastasia Makarova and Andreas Krause},
url = {https://openreview.net/forum?id=YcW8i9VCf5},
year = {2024},
date = {2024-01-01},
booktitle = {The Twelfth International Conference on Learning Representations},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Anani, Alaa; Lorenz, Tobias; Schiele, Bernt; Fritz, Mario
Adaptive hierarchical certification for segmentation using randomized smoothing Proceedings Article
In: Proceedings of the 41st International Conference on Machine Learning, JMLR.org, Vienna, Austria, 2024.
@inproceedings{10.5555/3692070.3692132,
title = {Adaptive hierarchical certification for segmentation using randomized smoothing},
author = {Alaa Anani and Tobias Lorenz and Bernt Schiele and Mario Fritz},
year = {2024},
date = {2024-01-01},
booktitle = {Proceedings of the 41st International Conference on Machine Learning},
publisher = {JMLR.org},
address = {Vienna, Austria},
series = {ICML'24},
abstract = {Certification for machine learning is proving that no adversarial sample can evade a model within a range under certain conditions, a necessity for safety-critical domains. Common certification methods for segmentation use a flat set of fine-grained classes, leading to high abstain rates due to model uncertainty across many classes. We propose a novel, more practical setting, which certifies pixels within a multi-level hierarchy, and adaptively relaxes the certification to a coarser level for unstable components classic methods would abstain from, effectively lowering the abstain rate whilst providing more certified semantically meaningful information. We mathematically formulate the problem setup, introduce an adaptive hierarchical certification algorithm and prove the correctness of its guarantees. Since certified accuracy does not take the loss of information into account for coarser classes, we introduce the Certified Information Gain (CIG) metric, which is proportional to the class granularity level. Our extensive experiments on the datasets Cityscapes, PASCAL-Context, ACDC and COCO-Stuff demonstrate that our adaptive algorithm achieves a higher CIG and lower abstain rate compared to the current state-of-the-art certification method. Our code can be found here: https://github.com/AlaaAnani/adaptive-certify.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Certification for machine learning is proving that no adversarial sample can evade a model within a range under certain conditions, a necessity for safety-critical domains. Common certification methods for segmentation use a flat set of fine-grained classes, leading to high abstain rates due to model uncertainty across many classes. We propose a novel, more practical setting, which certifies pixels within a multi-level hierarchy, and adaptively relaxes the certification to a coarser level for unstable components classic methods would abstain from, effectively lowering the abstain rate whilst providing more certified semantically meaningful information. We mathematically formulate the problem setup, introduce an adaptive hierarchical certification algorithm and prove the correctness of its guarantees. Since certified accuracy does not take the loss of information into account for coarser classes, we introduce the Certified Information Gain (CIG) metric, which is proportional to the class granularity level. Our extensive experiments on the datasets Cityscapes, PASCAL-Context, ACDC and COCO-Stuff demonstrate that our adaptive algorithm achieves a higher CIG and lower abstain rate compared to the current state-of-the-art certification method. Our code can be found here: https://github.com/AlaaAnani/adaptive-certify.
Chen, Dingfan; Kerkouche, Raouf; Fritz, Mario
A Unified View of Differentially Private Deep Generative Modeling Journal Article
In: Transactions on Machine Learning Research, 2024, ISSN: 2835-8856, (Survey Certification).
@article{chen2024a,
title = {A Unified View of Differentially Private Deep Generative Modeling},
author = {Dingfan Chen and Raouf Kerkouche and Mario Fritz},
url = {https://openreview.net/forum?id=YgmBD2c9qX},
issn = {2835-8856},
year = {2024},
date = {2024-01-01},
journal = {Transactions on Machine Learning Research},
note = {Survey Certification},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Räisä, Ossi; Honkela, Antti
A Bias-Variance Decomposition for Ensembles over Multiple Synthetic Datasets Journal Article
In: ArXiv, vol. abs/2402.03985, 2024.
@article{Ris2024ABD,
title = {A Bias-Variance Decomposition for Ensembles over Multiple Synthetic Datasets},
author = {Ossi Räisä and Antti Honkela},
url = {https://api.semanticscholar.org/CorpusID:267500003},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {ArXiv},
volume = {abs/2402.03985},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Ogun, Sewade; Owodunni, Abraham T.; Olatunji, Tobi; Alese, Eniola; Oladimeji, Babatunde; Afonja, Tejumade; Olaleye, Kayode; Etori, Naome A.; Adewumi, Tosin
1000 African Voices: Advancing inclusive multi-speaker multi-accent speech synthesis Proceedings Article
In: Interspeech 2024, pp. 1855–1859, 2024, ISSN: 2958-1796.
@inproceedings{ogun24_interspeech,
title = {1000 African Voices: Advancing inclusive multi-speaker multi-accent speech synthesis},
author = {Sewade Ogun and Abraham T. Owodunni and Tobi Olatunji and Eniola Alese and Babatunde Oladimeji and Tejumade Afonja and Kayode Olaleye and Naome A. Etori and Tosin Adewumi},
doi = {10.21437/Interspeech.2024-2281},
issn = {2958-1796},
year = {2024},
date = {2024-01-01},
booktitle = {Interspeech 2024},
pages = {1855–1859},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Gibert, Daniel; Demetrio, Luca; Zizzo, Giulio; Le, Quan; Planes, Jordi; Biggio, Battista
Certified Adversarial Robustness of Machine Learning-based Malware Detectors via (De)Randomized Smoothing Journal Article
In: arXiv preprint arXiv:2405.00392, 2024.
@article{gibert24-arxiv,
title = {Certified Adversarial Robustness of Machine Learning-based Malware Detectors via (De)Randomized Smoothing},
author = {Daniel Gibert and Luca Demetrio and Giulio Zizzo and Quan Le and Jordi Planes and Battista Biggio},
year = {2024},
date = {2024-01-01},
journal = {arXiv preprint arXiv:2405.00392},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Demetrio, Luca; Biggio, Battista
secml-malware: Pentesting Windows Malware Classifiers with Adversarial EXEmples in Python Journal Article
In: ArXiv e-prints, 2024.
@article{demetrio24-secml,
title = {secml-malware: Pentesting Windows Malware Classifiers with Adversarial EXEmples in Python},
author = {Luca Demetrio and Battista Biggio},
year = {2024},
date = {2024-01-01},
journal = {ArXiv e-prints},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Guo, Wei; Demontis, Ambra; Pintor, Maura; Chan, Patrick P. K.; Biggio, Battista
LFPD: Local-Feature-Powered Defense Against Adaptive Backdoor Attacks Proceedings Article
In: International Conference on Machine Learning and Cybernetics, pp. 607-612, IEEE, 2024.
@inproceedings{guo24-icmlc,
title = {LFPD: Local-Feature-Powered Defense Against Adaptive Backdoor Attacks},
author = {Wei Guo and Ambra Demontis and Maura Pintor and Patrick P. K. Chan and Battista Biggio},
doi = {10.1109/ICMLC63072.2024.10935153},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {International Conference on Machine Learning and Cybernetics},
pages = {607-612},
publisher = {IEEE},
series = {ICMLC},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Minnei, Luca; Eddoubi, Hicham; Sotgiu, Angelo; Pintor, Maura; Demontis, Ambra; Biggio, Battista
Data Drift in Android Malware Detection Proceedings Article
In: International Conference on Machine Learning and Cybernetics, IEEE, 2024.
@inproceedings{minnei24-icmlc,
title = {Data Drift in Android Malware Detection},
author = {Luca Minnei and Hicham Eddoubi and Angelo Sotgiu and Maura Pintor and Ambra Demontis and Battista Biggio},
url = {https://ieeexplore.ieee.org/abstract/document/10935015},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {International Conference on Machine Learning and Cybernetics},
publisher = {IEEE},
series = {ICMLC},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
2023
Li, Yi; Angelov, Plamen; Suri, Neeraj
Fuzzy Detectors Against Adversarial Attacks Proceedings Article
In: 2023 IEEE Symposium Series on Computational Intelligence (SSCI), pp. 306-311, 2023, ISSN: 2472-8322.
@inproceedings{10372061,
title = {Fuzzy Detectors Against Adversarial Attacks},
author = {Yi Li and Plamen Angelov and Neeraj Suri},
doi = {10.1109/SSCI52147.2023.10372061},
issn = {2472-8322},
year = {2023},
date = {2023-12-01},
booktitle = {2023 IEEE Symposium Series on Computational Intelligence (SSCI)},
pages = {306-311},
abstract = {Deep learning-based methods have proved useful for adversarial attack detection. However, conventional detection algorithms exploit crisp set theory for classification boundary. Therefore, representing vague concepts is not available. Mo-tivated by the recent success in fuzzy systems, we propose a fuzzy rule-based neural network to improve adversarial attack detection accuracy. The pretrained ImageNet model is exploited to extract feature maps from clean and attacked images. Subsequently, the fuzzification network is used to obtain feature maps to produce fuzzy sets of difference degrees between clean and attacked images. The fuzzy rules control the intelligence that determines the detection boundaries. In the defuzzification layer, the fuzzy prediction from the in-telligence is mapped back into the crisp model predictions for images. The loss between the prediction and label controls the rules to train the fuzzy detector. We show that the fuzzy rule-based network learns rich feature information than binary outputs and offers to obtain an overall performance gain. Our experiments, conducted over a wide range of images, show that the proposed method consistently performs better than conventional crisp set training in adversarial attack detection with various fuzzy system-based neural networks. The source code of the proposed method is available at https://github.com/Yukino-3/Fuzzy.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Deep learning-based methods have proved useful for adversarial attack detection. However, conventional detection algorithms exploit crisp set theory for classification boundary. Therefore, representing vague concepts is not available. Mo-tivated by the recent success in fuzzy systems, we propose a fuzzy rule-based neural network to improve adversarial attack detection accuracy. The pretrained ImageNet model is exploited to extract feature maps from clean and attacked images. Subsequently, the fuzzification network is used to obtain feature maps to produce fuzzy sets of difference degrees between clean and attacked images. The fuzzy rules control the intelligence that determines the detection boundaries. In the defuzzification layer, the fuzzy prediction from the in-telligence is mapped back into the crisp model predictions for images. The loss between the prediction and label controls the rules to train the fuzzy detector. We show that the fuzzy rule-based network learns rich feature information than binary outputs and offers to obtain an overall performance gain. Our experiments, conducted over a wide range of images, show that the proposed method consistently performs better than conventional crisp set training in adversarial attack detection with various fuzzy system-based neural networks. The source code of the proposed method is available at https://github.com/Yukino-3/Fuzzy.
Rai, Shyam Nandan; Cermelli, Fabio; Fontanel, Dario; Masone, Carlo; Caputo, Barbara
Unmasking Anomalies in Road-Scene Segmentation Proceedings Article
In: 2023 IEEE/CVF International Conference on Computer Vision (ICCV), pp. 4014-4023, IEEE Computer Society, Los Alamitos, CA, USA, 2023.
@inproceedings{10378244,
title = { Unmasking Anomalies in Road-Scene Segmentation },
author = {Shyam Nandan Rai and Fabio Cermelli and Dario Fontanel and Carlo Masone and Barbara Caputo},
url = {https://doi.ieeecomputersociety.org/10.1109/ICCV51070.2023.00373},
doi = {10.1109/ICCV51070.2023.00373},
year = {2023},
date = {2023-10-01},
booktitle = {2023 IEEE/CVF International Conference on Computer Vision (ICCV)},
pages = {4014-4023},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
abstract = {Anomaly segmentation is a critical task for driving applications, and it is approached traditionally as a per-pixel classification problem. However, reasoning individually about each pixel without considering their contextual semantics results in high uncertainty around the objects’ boundaries and numerous false positives. We propose a paradigm change by shifting from a per-pixel classification to a mask classification. Our mask-based method, Mask2Anomaly, demonstrates the feasibility of integrating an anomaly detection method in a mask-classification architecture. Mask2Anomaly includes several technical novelties that are designed to improve the detection of anomalies in masks: i) a global masked attention module to focus individually on the foreground and background regions; ii) a mask contrastive learning that maximizes the margin between an anomaly and known classes; and iii) a mask refinement solution to reduce false positives. Mask2Anomaly achieves new state-of-the-art results across a range of benchmarks, both in the per-pixel and component-level evaluations. In particular, Mask2Anomaly reduces the average false positives rate by 60% w.r.t. the previous state-of-the-art.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Anomaly segmentation is a critical task for driving applications, and it is approached traditionally as a per-pixel classification problem. However, reasoning individually about each pixel without considering their contextual semantics results in high uncertainty around the objects’ boundaries and numerous false positives. We propose a paradigm change by shifting from a per-pixel classification to a mask classification. Our mask-based method, Mask2Anomaly, demonstrates the feasibility of integrating an anomaly detection method in a mask-classification architecture. Mask2Anomaly includes several technical novelties that are designed to improve the detection of anomalies in masks: i) a global masked attention module to focus individually on the foreground and background regions; ii) a mask contrastive learning that maximizes the margin between an anomaly and known classes; and iii) a mask refinement solution to reduce false positives. Mask2Anomaly achieves new state-of-the-art results across a range of benchmarks, both in the per-pixel and component-level evaluations. In particular, Mask2Anomaly reduces the average false positives rate by 60% w.r.t. the previous state-of-the-art.
Fahes, Mohammad; Vu, Tuan-Hung; Bursuc, Andrei; Pérez, Patrick; Charette, Raoul De
PØDA: Prompt-driven Zero-shot Domain Adaptation Proceedings Article
In: 2023 IEEE/CVF International Conference on Computer Vision (ICCV), pp. 18577-18587, 2023, ISSN: 2380-7504.
@inproceedings{10376773,
title = {PØDA: Prompt-driven Zero-shot Domain Adaptation},
author = {Mohammad Fahes and Tuan-Hung Vu and Andrei Bursuc and Patrick Pérez and Raoul De Charette},
doi = {10.1109/ICCV51070.2023.01707},
issn = {2380-7504},
year = {2023},
date = {2023-10-01},
booktitle = {2023 IEEE/CVF International Conference on Computer Vision (ICCV)},
pages = {18577-18587},
abstract = {Domain adaptation has been vastly investigated in computer vision but still requires access to target images at train time, which might be intractable in some uncommon conditions. In this paper, we propose the task of ‘Prompt-driven Zero-shot Domain Adaptation’, where we adapt a model trained on a source domain using only a general description in natural language of the target domain, i.e., a prompt. First, we leverage a pretrained contrastive vision-language model (CLIP) to optimize affine transformations of source features, steering them towards the target text embedding while preserving their content and semantics. To achieve this, we propose Prompt-driven Instance Normalization (PIN). Second, we show that these prompt-driven augmentations can be used to perform zero-shot domain adaptation for semantic segmentation. Experiments demonstrate that our method significantly outperforms CLIP-based style transfer baselines on several datasets for the downstream task at hand, even surpassing one-shot unsupervised domain adaptation. A similar boost is observed on object detection and image classification. The code is available at https://github.com/astra-vision/PODA.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Domain adaptation has been vastly investigated in computer vision but still requires access to target images at train time, which might be intractable in some uncommon conditions. In this paper, we propose the task of ‘Prompt-driven Zero-shot Domain Adaptation’, where we adapt a model trained on a source domain using only a general description in natural language of the target domain, i.e., a prompt. First, we leverage a pretrained contrastive vision-language model (CLIP) to optimize affine transformations of source features, steering them towards the target text embedding while preserving their content and semantics. To achieve this, we propose Prompt-driven Instance Normalization (PIN). Second, we show that these prompt-driven augmentations can be used to perform zero-shot domain adaptation for semantic segmentation. Experiments demonstrate that our method significantly outperforms CLIP-based style transfer baselines on several datasets for the downstream task at hand, even surpassing one-shot unsupervised domain adaptation. A similar boost is observed on object detection and image classification. The code is available at https://github.com/astra-vision/PODA.
Aghasanli, Agil; Kangin, Dmitry; Angelov, Plamen
Interpretable-through-prototypes deepfake detection for diffusion models Proceedings Article
In: 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), pp. 467-474, 2023, ISSN: 2473-9944.
@inproceedings{10350382,
title = {Interpretable-through-prototypes deepfake detection for diffusion models},
author = {Agil Aghasanli and Dmitry Kangin and Plamen Angelov},
doi = {10.1109/ICCVW60793.2023.00053},
issn = {2473-9944},
year = {2023},
date = {2023-10-01},
booktitle = {2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)},
pages = {467-474},
abstract = {The process of recognizing and distinguishing between real content and content generated by deep learning algorithms, often referred to as deepfakes, is known as deepfake detection. In order to counter the rising threat of deepfakes and maintain the integrity of digital media, research is now being done to create more reliable and precise detection techniques. Deep learning models, such as Stable Diffusion, have been able to generate more detailed and less blurry images in recent years. In this paper, we develop a deepfake detection technique to distinguish original and fake images generated by various Diffusion Models. The developed methodology for deepfake detection takes advantage of features from fine-tuned Vision Transformers (ViTs), combined with existing classifiers such as Support Vector Machines (SVM). We demonstrate the proposed methodology’s ability of interpretability-through-prototypes by analysing support vectors of the SVMs. Additionally, due to the novelty of the topic, there is a lack of open datasets for deepfake detection. Therefore, to evaluate the methodology, we have also created custom datasets based on various generative techniques of Diffusion Models on open datasets (ImageNet, FFHQ, Oxford-IIIT Pet). The code is available at https://github.com/lira-centre/DeepfakeDetection.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The process of recognizing and distinguishing between real content and content generated by deep learning algorithms, often referred to as deepfakes, is known as deepfake detection. In order to counter the rising threat of deepfakes and maintain the integrity of digital media, research is now being done to create more reliable and precise detection techniques. Deep learning models, such as Stable Diffusion, have been able to generate more detailed and less blurry images in recent years. In this paper, we develop a deepfake detection technique to distinguish original and fake images generated by various Diffusion Models. The developed methodology for deepfake detection takes advantage of features from fine-tuned Vision Transformers (ViTs), combined with existing classifiers such as Support Vector Machines (SVM). We demonstrate the proposed methodology’s ability of interpretability-through-prototypes by analysing support vectors of the SVMs. Additionally, due to the novelty of the topic, there is a lack of open datasets for deepfake detection. Therefore, to evaluate the methodology, we have also created custom datasets based on various generative techniques of Diffusion Models on open datasets (ImageNet, FFHQ, Oxford-IIIT Pet). The code is available at https://github.com/lira-centre/DeepfakeDetection.
Galli, Filippo; Jung, Kangsoo; Biswas, Sayan; Palamidessi, Catuscia; Cucinotta, Tommaso
Advancing Personalized Federated Learning: Group Privacy, Fairness, and Beyond Journal Article
In: SN Computer Science, vol. 4, no. 6, pp. 831, 2023.
@article{Galli2023,
title = {Advancing Personalized Federated Learning: Group Privacy, Fairness, and Beyond},
author = {Filippo Galli and Kangsoo Jung and Sayan Biswas and Catuscia Palamidessi and Tommaso Cucinotta},
url = {https://doi.org/10.1007/s42979-023-02292-0},
doi = {10.1007/s42979-023-02292-0},
year = {2023},
date = {2023-10-01},
journal = {SN Computer Science},
volume = {4},
number = {6},
pages = {831},
abstract = {Federated learning (FL) is a framework for training machine learning models in a distributed and collaborative manner. During training, a set of participating clients process their data stored locally, sharing only updates of the statistical model’s parameters obtained by minimizing a cost function over their local inputs. FL was proposed as a stepping-stone towards privacy-preserving machine learning, but it has been shown to expose clients to issues such as leakage of private information, lack of personalization of the model, and the possibility of having a trained model that is fairer to some groups of clients than to others. In this paper, the focus is on addressing the triadic interaction among personalization, privacy guarantees, and fairness attained by trained models within the FL framework. Differential privacy and its variants have been studied and applied as cutting-edge standards for providing formal privacy guarantees. However, clients in FL often hold very diverse datasets representing heterogeneous communities, making it important to protect their sensitive and personal information while still ensuring that the trained model upholds the aspect of fairness for the users. To attain this objective, a method is put forth that introduces group privacy assurances through the utilization of d-privacy (aka metric privacy). d-privacy represents a localized form of differential privacy that relies on a metric-oriented obfuscation approach to maintain the original data’s topological distribution. This method, besides enabling personalized model training in a federated approach and providing formal privacy guarantees, possesses significantly better group fairness measured under a variety of standard metrics than a global model trained within a classical FL template. Theoretical justifications for the applicability are provided, as well as experimental validation on real-world datasets to illustrate the working of the proposed method.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Federated learning (FL) is a framework for training machine learning models in a distributed and collaborative manner. During training, a set of participating clients process their data stored locally, sharing only updates of the statistical model’s parameters obtained by minimizing a cost function over their local inputs. FL was proposed as a stepping-stone towards privacy-preserving machine learning, but it has been shown to expose clients to issues such as leakage of private information, lack of personalization of the model, and the possibility of having a trained model that is fairer to some groups of clients than to others. In this paper, the focus is on addressing the triadic interaction among personalization, privacy guarantees, and fairness attained by trained models within the FL framework. Differential privacy and its variants have been studied and applied as cutting-edge standards for providing formal privacy guarantees. However, clients in FL often hold very diverse datasets representing heterogeneous communities, making it important to protect their sensitive and personal information while still ensuring that the trained model upholds the aspect of fairness for the users. To attain this objective, a method is put forth that introduces group privacy assurances through the utilization of d-privacy (aka metric privacy). d-privacy represents a localized form of differential privacy that relies on a metric-oriented obfuscation approach to maintain the original data’s topological distribution. This method, besides enabling personalized model training in a federated approach and providing formal privacy guarantees, possesses significantly better group fairness measured under a variety of standard metrics than a global model trained within a classical FL template. Theoretical justifications for the applicability are provided, as well as experimental validation on real-world datasets to illustrate the working of the proposed method.
Ledda, Emanuele; Angioni, Daniele; Piras, Giorgio; Fumera, Giorgio; Biggio, Battista; Roli, Fabio
Adversarial Attacks Against Uncertainty Quantification Proceedings Article
In: 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), pp. 4601-4610, 2023.
@inproceedings{ledda23-iccvw,
title = {Adversarial Attacks Against Uncertainty Quantification},
author = {Emanuele Ledda and Daniele Angioni and Giorgio Piras and Giorgio Fumera and Battista Biggio and Fabio Roli},
year = {2023},
date = {2023-10-01},
booktitle = {2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)},
pages = {4601-4610},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Li, Zheng; Yu, Ning; Salem, Ahmed; Backes, Michael; Fritz, Mario; Zhang, Yang
UnGANable: Defending Against GAN-based Face Manipulation Proceedings Article
In: 32nd USENIX Security Symposium (USENIX Security 23), pp. 7213–7230, USENIX Association, Anaheim, CA, 2023, ISBN: 978-1-939133-37-3.
@inproceedings{285407,
title = {UnGANable: Defending Against GAN-based Face Manipulation},
author = {Zheng Li and Ning Yu and Ahmed Salem and Michael Backes and Mario Fritz and Yang Zhang},
url = {https://www.usenix.org/conference/usenixsecurity23/presentation/li-zheng},
isbn = {978-1-939133-37-3},
year = {2023},
date = {2023-08-01},
booktitle = {32nd USENIX Security Symposium (USENIX Security 23)},
pages = {7213–7230},
publisher = {USENIX Association},
address = {Anaheim, CA},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Jiang, Ziping; Wang, Yunpeng; Li, Chang-Tsun; Angelov, Plamen; Jiang, Richard
Delve Into Neural Activations: Toward Understanding Dying Neurons Journal Article
In: IEEE Transactions on Artificial Intelligence, vol. 4, no. 4, pp. 959-971, 2023, ISSN: 2691-4581.
@article{9792293,
title = {Delve Into Neural Activations: Toward Understanding Dying Neurons},
author = {Ziping Jiang and Yunpeng Wang and Chang-Tsun Li and Plamen Angelov and Richard Jiang},
doi = {10.1109/TAI.2022.3180272},
issn = {2691-4581},
year = {2023},
date = {2023-08-01},
journal = {IEEE Transactions on Artificial Intelligence},
volume = {4},
number = {4},
pages = {959-971},
abstract = {Theoretically, a deep neuron network with nonlinear activation is able to approximate any function, while empirically the performance of the model with different activations varies widely. In this work, we investigate the expressivity of the network from an activation perspective. In particular, we introduce a generalized activation region/pattern to describe the functional relationship of the model with an arbitrary activation function and illustrate its fundamental properties. We then propose a metric named pattern similarity to evaluate the practical expressivity of neuron networks regarding datasets based on the neuron level reaction toward the input. We find an undocumented dying neuron issue that the postactivation value of most neurons remain in the same region for data with different labels, implying that the expressivity of the network with certain activations is greatly constrained. For instance, around 80% of postactivation values of a well-trained Sigmoid net or Tanh net are clustered in the same region given any test sample. This means most of the neurons fail to provide any useful information in distinguishing the data with different labels, suggesting that the practical expressivity of those networks is far below the theoretical. By evaluating our metrics and the test accuracy of the model, we show that the seriousness of the dying neuron issue is highly related to the model performance. At last, we also discussed the cause of the dying neuron issue, providing an explanation of the model performance gap caused by the choice of activation.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Theoretically, a deep neuron network with nonlinear activation is able to approximate any function, while empirically the performance of the model with different activations varies widely. In this work, we investigate the expressivity of the network from an activation perspective. In particular, we introduce a generalized activation region/pattern to describe the functional relationship of the model with an arbitrary activation function and illustrate its fundamental properties. We then propose a metric named pattern similarity to evaluate the practical expressivity of neuron networks regarding datasets based on the neuron level reaction toward the input. We find an undocumented dying neuron issue that the postactivation value of most neurons remain in the same region for data with different labels, implying that the expressivity of the network with certain activations is greatly constrained. For instance, around 80% of postactivation values of a well-trained Sigmoid net or Tanh net are clustered in the same region given any test sample. This means most of the neurons fail to provide any useful information in distinguishing the data with different labels, suggesting that the practical expressivity of those networks is far below the theoretical. By evaluating our metrics and the test accuracy of the model, we show that the seriousness of the dying neuron issue is highly related to the model performance. At last, we also discussed the cause of the dying neuron issue, providing an explanation of the model performance gap caused by the choice of activation.