Publications

Evaluating the effectiveness of the Smart About Meds (SAM) mobile application among patients discharged from hospital: protocol of a randomised controlled trial
Robyn Tamblyn
Bettina Habib
David L Buckeridge
Daniala L Weir
Elizaveta Frolova
Rolan Alattar
Jessica Rogozinsky
Caroline Beauchamp
Rosalba Pupo
Susan J Bartlett
Emily McDonald
NCT05371548.
2024-10-31
BMJ Open (published)
doi.org
Hessian Aware Low-Rank Perturbation for Order-Robust Continual Learning
Jiaqi Li
Rui Wang
Yuanhao Lai
Changjian Shui
Sabyasachi Sahoo
Charles X. Ling
Shichun Yang
Boyu Wang
Christian Gagné
Fan Zhou
Continual learning aims to learn a series of tasks sequentially without forgetting the knowledge acquired from the previous ones. In this wo… (see more)rk, we propose the Hessian Aware Low-Rank Perturbation algorithm for continual learning. By modeling the parameter transitions along the sequential tasks with the weight matrix transformation, we propose to apply the low-rank approximation on the task-adaptive parameters in each layer of the neural networks. Specifically, we theoretically demonstrate the quantitative relationship between the Hessian and the proposed low-rank approximation. The approximation ranks are then globally determined according to the marginal increment of the empirical loss estimated by the layer-specific gradient and low-rank approximation error. Furthermore, we control the model capacity by pruning less important parameters to diminish the parameter growth. We conduct extensive experiments on various benchmarks, including a dataset with large-scale tasks, and compare our method against some recent state-of-the-art methods to demonstrate the effectiveness and scalability of our proposed method. Empirical results show that our method performs better on different benchmarks, especially in achieving task order robustness and handling the forgetting issue. The source code is at https://github.com/lijiaqi/HALRP.
2024-10-31
IEEE Transactions on Knowledge and Data Engineering (published)
doi.org
arxiv.org
Improving Adversarial Robustness in Vision-Language Models with Architecture and Prompt Design.
Rishika Bhagwatkar
Shravan Nayak
Pouya Bashivan
Irina Rish
2024-10-31
Findings of the Association for Computational Linguistics: EMNLP 2024 (published)
doi.org
Inhibition of the inferior parietal lobe triggers state-dependent network adaptations
Kathleen A. Williams
Ole Numssen
Juan David Guerra
Jakub Kopal
Danilo Bzdok
Gesa Hartwigsen
The human brain comprises large-scale networks that flexibly interact to support diverse cognitive functions and adapt to variability in dai… (see more)ly life. The inferior parietal lobe (IPL) is a hub of multiple brain networks that sustain various cognitive domains. It remains unclear how networks respond to acute regional perturbations to maintain normal function. To provoke network-level adaptive responses to local inhibition, we combined offline transcranial magnetic stimulation (TMS) over left or right IPL with neuroimaging during attention, semantic and social cognition tasks, and rest. Across tasks, TMS specifically affected task-active network activity with inhibition and facilitation. Network interaction responses differed between rest and tasks. After TMS over both IPL regions, large-scale network interactions were exclusively facilitated at rest, but mainly inhibited during tasks. Overall, responses to TMS primarily occurred in and between domain-general default mode and frontoparietal subnetworks. These findings elucidate short-term adaptive plasticity in response to network node inhibition.
2024-10-31
Heliyon (published)
doi.org
Do Large Language Models Know How Much They Know?
Gabriele Prato
Jerry Huang
Prasanna Parthasarathi
Shagun Sodhani
A. Chandar
2024-10-31
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (published)
doi.org
arxiv.org
OpenQDC: Open Quantum Data Commons
Cristian Gabellini
Nikhil Shenoy
Stephan Thaler
Semih Cantürk
Daniel McNeela
Dominique Beaini
Michael Bronstein
Prudencio Tossou
Machine Learning Interatomic Potentials (MLIPs) are a highly promising alternative to force-fields for molecular dynamics (MD) simulations, … (see more)offering precise and rapid energy and force calculations. However, Quantum-Mechanical (QM) datasets, crucial for MLIPs, are fragmented across various repositories, hindering accessibility and model development. We introduce the openQDC package, consolidating 37 QM datasets from over 250 quantum methods and 400 million geometries into a single, accessible resource. These datasets are meticulously preprocessed, and standardized for MLIP training, covering a wide range of chemical elements and interactions relevant in organic chemistry. OpenQDC includes tools for normalization and integration, easily accessible via Python. Experiments with well-known architectures like SchNet, TorchMD-Net, and DimeNet reveal challenges for those architectures and constitute a leaderboard to accelerate benchmarking and guide novel algorithms development. Continuously adding datasets to OpenQDC will democratize QM dataset access, foster more collaboration and innovation, enhance MLIP development, and support their adoption in the MD field.
2024-10-31
arXiv (published)
doi.org
Predictive Modeling of Body Image Dissatisfaction in People With Type 1 Diabetes
COURTNEY SOUTH
SHAHRYAR EBRAHIMI
ANNE-SOPHIE BRAZEAU
Maria Cutumisu
2024-10-31
Canadian Journal of Diabetes (published)
doi.org
SpeechBrain-MOABB: An open-source Python library for benchmarking deep neural networks applied to EEG signals
Davide Borra
Francesco Paissan
Mirco Ravanelli
2024-10-31
Computers in Biology and Medicine (published)
doi.org
Improving Adversarial Transferability via Model Alignment
Avery Ma
Amir-massoud Farahmand
Yangchen Pan
Philip Torr
Jindong Gu
Neural networks are susceptible to adversarial perturbations that are transferable across different models. In this paper, we introduce a no… (see more)vel model alignment technique aimed at improving a given source model's ability in generating transferable adversarial perturbations. During the alignment process, the parameters of the source model are fine-tuned to minimize an alignment loss. This loss measures the divergence in the predictions between the source model and another, independently trained model, referred to as the witness model. To understand the effect of model alignment, we conduct a geometric analysis of the resulting changes in the loss landscape. Extensive experiments on the ImageNet dataset, using a variety of model architectures, demonstrate that perturbations generated from aligned source models exhibit significantly higher transferability than those from the original source model.
2024-10-30
Lecture Notes in Computer Science (published)
doi.org
arxiv.org
Soft Condorcet Optimization for Ranking of General Agents
Marc Lanctot
Kate Larson
Michael Kaisers
Quentin Berthet
Ian Gemp
Manfred Diaz
Roberto-Rafael Maura-Rivero
Yoram Bachrach
Anna Koop
Doina Precup
A common way to drive progress of AI models and agents is to compare their performance on standardized benchmarks. Comparing the performance… (see more) of general agents requires aggregating their individual performances across a potentially wide variety of different tasks. In this paper, we describe a novel ranking scheme inspired by social choice frameworks, called Soft Condorcet Optimization (SCO), to compute the optimal ranking of agents: the one that makes the fewest mistakes in predicting the agent comparisons in the evaluation data. This optimal ranking is the maximum likelihood estimate when evaluation data (which we view as votes) are interpreted as noisy samples from a ground truth ranking, a solution to Condorcet's original voting system criteria. SCO ratings are maximal for Condorcet winners when they exist, which we show is not necessarily true for the classical rating system Elo. We propose three optimization algorithms to compute SCO ratings and evaluate their empirical performance. When serving as an approximation to the Kemeny-Young voting method, SCO rankings are on average 0 to 0.043 away from the optimal ranking in normalized Kendall-tau distance across 865 preference profiles from the PrefLib open ranking archive. In a simulated noisy tournament setting, SCO achieves accurate approximations to the ground truth ranking and the best among several baselines when 59\% or more of the preference data is missing. Finally, SCO ranking provides the best approximation to the optimal ranking, measured on held-out test sets, in a problem containing 52,958 human players across 31,049 games of the classic seven-player game of Diplomacy.
2024-10-30
ArXiv (preprint)
doi.org
arxiv.org
Unclocklike biological oscillators with frequency memory
Christian Mauffette Denis
Paul François
Entrainment experiments on the vertebrate segmentation clock have revealed that embryonic oscillators actively change their internal frequen… (see more)cy to adapt to the driving signal. This is not consistent with either a one-dimensional clock model or a limit-cycle model, but rather it suggests a new “unclocklike” behavior. In this work, we propose simple, biologically realistic descriptions of such internal frequency adaptation, where a phase oscillator activates a memory variable controlling the oscillator's frequency. We study two opposite limits for the control of the memory variable, one with a smooth phase-averaging memory field, and the other with a pulsatile, phase-dependent activation. Both models recapitulate intriguing properties of the entrained segmentation clock, such as very broad Arnold tongues and an entrainment phase plateauing with detuning. We compute analytically multiple properties of such systems, such as entrainment phases and cycle shapes. We further describe new phenomena, including hysteresis in entrainment, bistability in the frequency of the entrained oscillator, and probabilistic entrainment. Our work shows that oscillators with frequency memory can exhibit new classes of unclocklike properties that can be tested through experimental entrainment. Published by the American Physical Society 2024
2024-10-30
Physical Review X Life (published)
doi.org
arxiv.org
Beyond Causal Discovery for Astronomy: Learning Meaningful Representations with Independent Component Analysis
Zehao Jin
Mario Pasquato
Benjamin L. Davis
Andrea Maccio
Yashar Hezaveh
2024-10-29
NeurIPS.cc/2024/Workshop/CRL (poster)
openreview.net
openreview.net