Publications

Neural representation of occluded objects in visual cortex
Courtney Mansfield
Tim Kietzmann
Jasper JF van den Bosch
Ian Charest
Marieke Mur
Nikolaus Kriegeskorte
Fraser Smith
2023-08-01
Journal of Vision (publié)
doi.org
Reconstructing mental images using Bubbles and electroencephalography
Audrey Lamy-Proulx
Jasper JF van den Bosch
Catherine Landry
Peter Brotherwood
Vincent Taschereau-Dumouchel
Frédéric Gosselin
Ian Charest
2023-08-01
Journal of Vision (publié)
doi.org
Scientific discovery in the age of artificial intelligence
Hanchen Wang
Tianfan Fu
Yuanqi Du
Wenhao Gao
Kexin Huang
Ziming Liu
Payal Chandak
Shengchao Liu
Peter Van Katwyk
Andreea Deac
Animashree Anandkumar
K. Bergen
Carla P. Gomes
Shirley Ho
Pushmeet Kohli
Joan Lasenby
Jure Leskovec
Tie-Yan Liu
A. Manrai
Debora Susan Marks … (voir 10 de plus)
Bharath Ramsundar
Le Song
Jimeng Sun
Jian Tang
Petar Veličković
Max Welling
Linfeng Zhang
Connor Wilson. Coley
Yoshua Bengio
Marinka Žitnik
2023-08-01
Nature (publié)
doi.org
The Different Faces of AI Ethics Across the World: A Principle-to-Practice Gap Analysis
Lionel Nganyewou Tidjon
Foutse Khomh
Artificial Intelligence (AI) is transforming our daily life with many applications in healthcare, space exploration, banking, and finance. T… (voir plus)his rapid progress in AI has brought increasing attention to the potential impacts of AI technologies on society, with ethically questionable consequences. In recent years, several ethical principles have been released by governments, national organizations, and international organizations. These principles outline high-level precepts to guide the ethical development, deployment, and governance of AI. However, the abstract nature, diversity, and context-dependence of these principles make them difficult to implement and operationalize, resulting in gaps between principles and their execution. Most recent work analyzed and summarized existing AI principles and guidelines but did not provide findings on principle-to-practice gaps nor how to mitigate them. These findings are particularly important to ensure that AI practical guidances are aligned with ethical principles and values. In this article, we provide a contextual and global evaluation of current ethical AI principles for all continents, with the aim to identify potential principle characteristics tailored to specific countries or applicable across countries. Next, we analyze the current level of AI readiness and current practical guidances of ethical AI principles in different countries, to identify gaps in the practical guidance of AI principles and their causes. Finally, we propose recommendations to mitigate the principle-to-practice gaps.
2023-08-01
IEEE Transactions on Artificial Intelligence (publié)
doi.org
The semantic distance between a linguistic prime and a natural scene target predicts reaction times in a visual search experiment
Katerina Marie Simkova
Jasper JF van den Bosch
Damiano Grignolio
Clayton Hickey
Ian Charest
2023-08-01
Journal of Vision (publié)
doi.org
Variational Nested Dropout
Yufei Cui
Yu Mao
Ziquan Liu
Qiao Li
Antoni B. Chan
Xue (Steve) Liu
Tei-Wei Kuo
Chun Jason Xue
Nested dropout is a variant of dropout operation that is able to order network parameters or features based on the pre-defined importance du… (voir plus)ring training. It has been explored for: I. Constructing nested nets Cui et al. 2020, Cui et al. 2021: the nested nets are neural networks whose architectures can be adjusted instantly during testing time, e.g., based on computational constraints. The nested dropout implicitly ranks the network parameters, generating a set of sub-networks such that any smaller sub-network forms the basis of a larger one. II. Learning ordered representation Rippel et al. 2014: the nested dropout applied to the latent representation of a generative model (e.g., auto-encoder) ranks the features, enforcing explicit order of the dense representation over dimensions. However, the dropout rate is fixed as a hyper-parameter during the whole training process. For nested nets, when network parameters are removed, the performance decays in a human-specified trajectory rather than in a trajectory learned from data. For generative models, the importance of features is specified as a constant vector, restraining the flexibility of representation learning. To address the problem, we focus on the probabilistic counterpart of the nested dropout. We propose a variational nested dropout (VND) operation that draws samples of multi-dimensional ordered masks at a low cost, providing useful gradients to the parameters of nested dropout. Based on this approach, we design a Bayesian nested neural network that learns the order knowledge of the parameter distributions. We further exploit the VND under different generative models for learning ordered latent distributions. In experiments, we show that the proposed approach outperforms the nested network in terms of accuracy, calibration, and out-of-domain detection in classification tasks. It also outperforms the related generative models on data generation tasks.
2023-08-01
IEEE Transactions on Pattern Analysis and Machine Intelligence (publié)
doi.org
arxiv.org
Do visual mental imagery and exteroceptive perception rely on the same mechanisms?
Catherine Landry
Jasper JF van den Bosch
Ian Charest
Frédéric Gosselin
Vincent Taschereau-Dumouchel
2023-08-01
Journal of Vision (publié)
doi.org
Evaluating Correctness and Faithfulness of Instruction-Following Models for Question Answering
Vaibhav Adlakha
Parishad BehnamGhader
Xing Han Lu
Nicholas Meade
Siva Reddy
Retriever-augmented instruction-following models are attractive alternatives to fine-tuned approaches for information-seeking tasks such as … (voir plus)question answering (QA). By simply prepending retrieved documents in its input along with an instruction, these models can be adapted to various information domains and tasks without additional fine-tuning. While the model responses tend to be natural and fluent, the additional verbosity makes traditional QA evaluation metrics such as exact match (EM) and F1 unreliable for accurately quantifying model performance. In this work, we investigate the performance of instruction-following models across three information-seeking QA tasks. We use both automatic and human evaluation to evaluate these models along two dimensions: 1) how well they satisfy the user's information need (correctness), and 2) whether they produce a response based on the provided knowledge (faithfulness). Guided by human evaluation and analysis, we highlight the shortcomings of traditional metrics for both correctness and faithfulness. We then propose simple token-overlap based and model-based metrics that reflect the true performance of these models. Our analysis reveals that instruction-following models are competitive, and sometimes even outperform fine-tuned models for correctness. However, these models struggle to stick to the provided knowledge and often hallucinate in their responses. We hope our work encourages a more holistic evaluation of instruction-following models for QA. Our code and data is available at https://github.com/McGill-NLP/instruct-qa
2023-07-31
ArXiv (prépublication)
doi.org
arxiv.org
GRouNdGAN: GRN-guided simulation of single-cell RNA-seq data using causal generative adversarial networks
Yazdan Zinati
Abdulrahman Takiddeen
Amin Emad
We introduce GRouNdGAN, a gene regulatory network (GRN)-guided causal implicit generative model for simulating single-cell RNA-seq data, in-… (voir plus)silico perturbation experiments, and benchmarking GRN inference methods. Through the imposition of a user-defined GRN in its architecture, GRouNdGAN simulates steady-state and transient-state single-cell datasets where genes are causally expressed under the control of their regulating transcription factors (TFs). Training on three experimental datasets, we show that our model captures non-linear TF-gene dependences and preserves gene identities, cell trajectories, pseudo-time ordering, and technical and biological noise, with no user manipulation and only implicit parameterization. Despite imposing rigid causality constraints, it outperforms state-of-the-art simulators in generating realistic cells. GRouNdGAN learns meaningful causal regulatory dynamics, allowing sampling from both observational and interventional distributions. This enables it to synthesize cells under conditions that do not occur in the dataset at inference time, allowing to perform in-silico TF knockout experiments. Our results show that in-silico knockout of cell type-specific TFs significantly reduces cells of that type being generated. Interactions imposed through the GRN are emphasized in the simulated datasets, resulting in GRN inference algorithms assigning them much higher scores than interactions not imposed but of equal importance in the experimental training dataset. Benchmarking various GRN inference algorithms reveals that GRouNdGAN effectively bridges the existing gap between simulated and biological data benchmarks of GRN inference algorithms, providing gold standard ground truth GRNs and realistic cells corresponding to the biological system of interest. Our results show that GRouNdGAN is a stable, realistic, and effective simulator with various applications in single-cell RNA-seq analysis.
2023-07-31
bioRxiv (prépublication)
doi.org
Multivariate analytical approaches for investigating brain-behavior relationships
E. Leighton Durham
Karam Ghanem
Andrew J. Stier
Carlos Cardenas-Iniguez
Gabrielle E. Reimann
Hee Jung Jeong
Randolph M. Dupont
Xiaoyu Dong
Tyler M. Moore
Marc G. Berman
Benjamin B. Lahey
Danilo Bzdok
Antonia N. Kaczkurkin
2023-07-31
Frontiers in Neuroscience (publié)
doi.org
FASHION AND SUSTAINABILITY: A SYSTEMATIC LITERATURE REVIEW
Osmud Rahman
Dingtao Hu
Benjamin Fung
2023-07-30
Global Fashion Management Conference (publié)
doi.org
GPS++: Reviving the Art of Message Passing for Molecular Property Prediction
Dominic Masters
Josef Dean
Kerstin Klaeser
Zhiyi Li
Samuel Maddrell-Mander
Adam Sanders
Hatem Helal
Deniz Beker
Andrew William Fitzgibbon
Shenyang Huang
Ladislav Rampášek
Dominique Beaini
2023-07-30
TMLR (accepté)
doi.org
openreview.net