Publications

Tackling the XAI Disagreement Problem with Regional Explanations
Gabriel Laberge
Yann Batiste Pequignot
Mario Marchand
Foutse Khomh
2024-04-17
Proceedings of The 27th International Conference on Artificial Intelligence and Statistics (published)
proceedings.mlr.press
proceedings.mlr.press
On the Privacy of Selection Mechanisms with Gaussian Noise
Jonathan Lebensold
Doina Precup
Borja Balle
Report Noisy Max and Above Threshold are two classical differentially private (DP) selection mechanisms. Their output is obtained by adding … (see more)noise to a sequence of low-sensitivity queries and reporting the identity of the query whose (noisy) answer satisfies a certain condition. Pure DP guarantees for these mechanisms are easy to obtain when Laplace noise is added to the queries. On the other hand, when instantiated using Gaussian noise, standard analyses only yield approximate DP guarantees despite the fact that the outputs of these mechanisms lie in a discrete space. In this work, we revisit the analysis of Report Noisy Max and Above Threshold with Gaussian noise and show that, under the additional assumption that the underlying queries are bounded, it is possible to provide pure ex-ante DP bounds for Report Noisy Max and pure ex-post DP bounds for Above Threshold. The resulting bounds are tight and depend on closed-form expressions that can be numerically evaluated using standard methods. Empirically we find these lead to tighter privacy accounting in the high privacy, low data regime. Further, we propose a simple privacy filter for composing pure ex-post DP guarantees, and use it to derive a fully adaptive Gaussian Sparse Vector Technique mechanism. Finally, we provide experiments on mobility and energy consumption datasets demonstrating that our Sparse Vector Technique is practically competitive with previous approaches and requires less hyper-parameter tuning.
2024-04-17
Proceedings of The 27th International Conference on Artificial Intelligence and Statistics (published)
doi.org
proceedings.mlr.press
Weight-Sharing Regularization
Mehran Shakerinava
Motahareh Sohrabi
Siamak Ravanbakhsh
Simon Lacoste-Julien
Weight-sharing is ubiquitous in deep learning. Motivated by this, we propose a "weight-sharing regularization" penalty on the weights …
2024-04-17
Proceedings of The 27th International Conference on Artificial Intelligence and Statistics (published)
doi.org
proceedings.mlr.press
Who Validates the Validators? Aligning LLM-Assisted Evaluation of LLM Outputs with Human Preferences
Shreya Shankar
J.D. Zamfirescu-Pereira
Bjorn Hartmann
Aditya G Parameswaran
Ian Arawjo
2024-04-17
ArXiv (preprint)
doi.org
arxiv.org
Asynchronous Algorithmic Alignment with Cocycles
Andrew Joseph Dudzik
Tamara von Glehn
Razvan Pascanu
Petar Veličković
State-of-the-art neural algorithmic reasoners make use of message passing in graph neural networks (GNNs). But typical GNNs blur the distinc… (see more)tion between the definition and invocation of the message function, forcing a node to send messages to its neighbours at every layer, synchronously. When applying GNNs to learn to execute dynamic programming algorithms, however, on most steps only a handful of the nodes would have meaningful updates to send. One, hence, runs the risk of inefficiencies by sending too much irrelevant data across the graph. But more importantly, many intermediate GNN steps have to learn the identity functions, which is a non-trivial learning problem. In this work, we explicitly separate the concepts of node state update and message function invocation. With this separation, we obtain a mathematical formulation that allows us to reason about asynchronous computation in both algorithms and neural networks. Our analysis yields several practical implementations of synchronous scalable GNN layers that are provably invariant under various forms of asynchrony.
2024-04-16
Proceedings of the Second Learning on Graphs Conference (published)
doi.org
proceedings.mlr.press
Latent Space Representations of Neural Algorithmic Reasoners
Vladimir V. Mirjani'c
Razvan Pascanu
Petar Velivckovi'c University of Cambridge
Petar Veličković
Google Deepmind
Neural Algorithmic Reasoning (NAR) is a research area focused on designing neural architectures that can reliably capture classical computat… (see more)ion, usually by learning to execute algorithms. A typical approach is to rely on Graph Neural Network (GNN) architectures, which encode inputs in high-dimensional latent spaces that are repeatedly transformed during the execution of the algorithm. In this work we perform a detailed analysis of the structure of the latent space induced by the GNN when executing algorithms. We identify two possible failure modes: (i) loss of resolution, making it hard to distinguish similar values; (ii) inability to deal with values outside the range observed during training. We propose to solve the first issue by relying on a softmax aggregator, and propose to decay the latent space in order to deal with out-of-range values. We show that these changes lead to improvements on the majority of algorithms in the standard CLRS-30 benchmark when using the state-of-the-art Triplet-GMPNN processor. Our code is available at https://github.com/mirjanic/nar-latent-spaces
2024-04-16
Proceedings of the Second Learning on Graphs Conference (published)
doi.org
proceedings.mlr.press
Government Interventions to Avert Future Catastrophic AI Risks
Yoshua Bengio
2024-04-14
Special Issue 5: Grappling With the Generative AI Revolution (published)
doi.org
Improving microbial phylogeny with citizen science within a mass-market video game
Roman Sarrazin-Gendron
Parham Ghasemloo Gheidari
Alexander Butyaev
Timothy Keding
Eddie Cai
Jiayue Zheng
Renata Mutalova
Julien Mounthanyvong
Yuxue Zhu
Elena Nazarova
Chrisostomos Drogaris
Kornél Erhart
David Michael Joshua Mathieu Vincent Steven Dan Jonathan Bélanger Bouffard Davidson Falaise Fiset Hebert He
David Michael Joshua Mathieu Vincent Steven Dan Jonathan Seung Jonathan David Steve Ludger Bélanger
David Bélanger
Michael Bouffard
Joshua Davidson
Mathieu Falaise
Vincent Fiset
Steven Hebert … (see 16 more)
Dan Hewitt
Jonathan Huot
Seung Kim
Jonathan Moreau-Genest
David Najjab
Steve Prince
Ludger Saintélien
Amélie Brouillette
Gabriel Richard
Randy Pitchford
Sébastien Caisse
Mathieu Blanchette
Daniel McDonald
Rob Knight
Attila Szantner
Jérôme Waldispühl
Citizen science video games are designed primarily for users already inclined to contribute to science, which severely limits their accessib… (see more)ility for an estimated community of 3 billion gamers worldwide. We created Borderlands Science (BLS), a citizen science activity that is seamlessly integrated within a popular commercial video game played by tens of millions of gamers. This integration is facilitated by a novel game-first design of citizen science games, in which the game design aspect has the highest priority, and a suitable task is then mapped to the game design. BLS crowdsources a multiple alignment task of 1 million 16S ribosomal RNA sequences obtained from human microbiome studies. Since its initial release on 7 April 2020, over 4 million players have solved more than 135 million science puzzles, a task unsolvable by a single individual. Leveraging these results, we show that our multiple sequence alignment simultaneously improves microbial phylogeny estimations and UniFrac effect sizes compared to state-of-the-art computational methods. This achievement demonstrates that hyper-gamified scientific tasks attract massive crowds of contributors and offers invaluable resources to the scientific community.
2024-04-14
Nature Biotechnology (published)
doi.org
Affirmative safety: An approach to risk management for high-risk AI
Akash Wasil
Joshua Clymer
David M. Krueger
Emily Dardaman
Simeon Campos
Evan Murphy
Prominent AI experts have suggested that companies developing high-risk AI systems should be required to show that such systems are safe bef… (see more)ore they can be developed or deployed. The goal of this paper is to expand on this idea and explore its implications for risk management. We argue that entities developing or deploying high-risk AI systems should be required to present evidence of affirmative safety: a proactive case that their activities keep risks below acceptable thresholds. We begin the paper by highlighting global security risks from AI that have been acknowledged by AI experts and world governments. Next, we briefly describe principles of risk management from other high-risk fields (e.g., nuclear safety). Then, we propose a risk management approach for advanced AI in which model developers must provide evidence that their activities keep certain risks below regulator-set thresholds. As a first step toward understanding what affirmative safety cases should include, we illustrate how certain kinds of technical evidence and operational evidence can support an affirmative safety case. In the technical section, we discuss behavioral evidence (evidence about model outputs), cognitive evidence (evidence about model internals), and developmental evidence (evidence about the training process). In the operational section, we offer examples of organizational practices that could contribute to affirmative safety cases: information security practices, safety culture, and emergency response capacity. Finally, we briefly compare our approach to the NIST AI Risk Management Framework. Overall, we hope our work contributes to ongoing discussions about national and global security risks posed by AI and regulatory approaches to address these risks.
2024-04-13
ArXiv (preprint)
doi.org
arxiv.org
CryCeleb: A Speaker Verification Dataset Based on Infant Cry Sounds
David Budaghyan
Arsenii Gorin
Cem Subakan
Charles C. Onu
Doina Precup
This paper describes the Ubenwa CryCeleb dataset - a labeled collection of infant cries - and the accompanying CryCeleb 2023 task, which is … (see more)a public speaker verification challenge based on cry sounds. We released more than 6 hours of manually segmented cry sounds from 786 newborns for academic use, aiming to encourage research in infant cry analysis. The inaugural public competition attracted 59 participants, 11 of whom improved the baseline performance. The top-performing system achieved a significant improvement scoring 25.8% equal error rate, which is still far from the performance of state-of-the-art adult speaker verification systems. Therefore, we believe there is room for further research on this dataset, potentially extending beyond the verification task.
2024-04-13
ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (published)
doi.org
arxiv.org
Directed Scattering for Knowledge Graph-Based Cellular Signaling Analysis
Aarthi Venkat
Joyce Chew
Ferran Cardoso Rodriguez
Christopher J. Tape
Michael Perlmutter
Smita Krishnaswamy
Directed graphs are a natural model for many phenomena, in particular scientific knowledge graphs such as molecular interaction or chemical … (see more)reaction networks that define cellular signaling relationships. In these situations, source nodes typically have distinct biophysical properties from sinks. Due to their ordered and unidirectional relationships, many such networks also have hierarchical and multiscale structure. However, the majority of methods performing node- and edge-level tasks in machine learning do not take these properties into account, and thus have not been leveraged effectively for scientific tasks such as cellular signaling network inference. We propose a new framework called Directed Scattering Autoencoder (DSAE) which uses a directed version of a geometric scattering transform, combined with the non-linear dimensionality reduction properties of an autoencoder and the geometric properties of the hyperbolic space to learn latent hierarchies. We show this method outperforms numerous others on tasks such as embedding directed graphs and learning cellular signaling networks.
2024-04-13
ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (published)
doi.org
arxiv.org
Focal Modulation Networks for Interpretable Sound Classification
Luca Della Libera
Cem Subakan
Mirco Ravanelli
The increasing success of deep neural networks has raised concerns about their inherent black-box nature, posing challenges related to inter… (see more)pretability and trust. While there has been extensive exploration of interpretation techniques in vision and language, interpretability in the audio domain has received limited attention, primarily focusing on post-hoc explanations. This paper addresses the problem of interpretability by-design in the audio domain by utilizing the recently proposed attention-free focal modulation networks (FocalNets). We apply FocalNets to the task of environmental sound classification for the first time and evaluate their interpretability properties on the popular ESC-50 dataset. Our method outperforms a similarly sized vision transformer both in terms of accuracy and interpretability. Furthermore, it is competitive against PIQ, a method specifically designed for post-hoc interpretation in the audio domain.
2024-04-13
2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW) (published)
doi.org
arxiv.org