Publications

Learning with Rejection for Abstractive Text Summarization

Meng Cao

Yue Dong

Jingyi He

Jackie CK Cheung

2021-12-31

EMNLP (published)

Local Structure Matters Most in Most Languages

Louis Clouatre

Prasanna Parthasarathi

Amal Zouaq

A. Chandar

2021-12-31

AACL/IJCNLP (2) (published)

Long Range Graph Benchmark

Vijay Prakash Dwivedi

Anh Tuan Luu

Graph Neural Networks (GNNs) that are based on the message passing (MP) paradigm generally exchange information between 1-hop neighbors to b… (see more)uild node representations at each layer. In principle, such networks are not able to capture long-range interactions (LRI) that may be desired or necessary for learning a given task on graphs. Recently, there has been an increasing interest in development of Transformer-based methods for graphs that can consider full node connectivity beyond the original sparse structure, thus enabling the modeling of LRI. However, MP-GNNs that simply rely on 1-hop message passing often fare better in several existing graph benchmarks when combined with positional feature representations, among other innovations, hence limiting the perceived utility and ranking of Transformer-like architectures. Here, we present the Long Range Graph Benchmark (LRGB) with 5 graph learning datasets: PascalVOC-SP, COCO-SP, PCQM-Contact, Peptides-func and Peptides-struct that arguably require LRI reasoning to achieve strong performance in a given task. We benchmark both baseline GNNs and Graph Transformer networks to verify that the models which capture long-range dependencies perform significantly better on these tasks. Therefore, these datasets are suitable for benchmarking and exploration of MP-GNNs and Graph Transformer architectures that are intended to capture LRI.

2021-12-31

Advances in Neural Information Processing Systems 35 (NeurIPS 2022) (published)

Low-Rank Representation of Reinforcement Learning Policies

Thang Doan

Guillaume Rabuseau

We propose a general framework for policy representation for reinforcement learning tasks. This framework involves finding a low-dimensional… (see more) embedding of the policy on a reproducing kernel Hilbert space (RKHS). The usage of RKHS based methods allows us to derive strong theoretical guarantees on the expected return of the reconstructed policy. Such guarantees are typically lacking in black-box models, but are very desirable in tasks requiring stability and convergence guarantees. We conduct several experiments on classic RL domains. The results confirm that the policies can be robustly represented in a low-dimensional space while the embedded policy incurs almost no decrease in returns.

2021-12-31

Journal of Artificial Intelligence Research (published)

Mastering the Unsupervised Reinforcement Learning Benchmark from Pixels

Sai Rajeswar

Pietro Mazzaglia

Tim Verbelen

Alexandre Piché

Bart Dhoedt

Aaron Courville

Alexandre Lacoste

Controlling artificial agents from visual sensory data is an arduous task. Reinforcement learning (RL) algorithms can succeed but require la… (see more)rge amounts of interactions between the agent and the environment. To alleviate the issue, unsupervised RL proposes to employ self-supervised interaction and learning, for adapting faster to future tasks. Yet, as shown in the Unsupervised RL Benchmark (URLB; Laskin et al. 2021), whether current unsupervised strategies can improve generalization capabilities is still unclear, especially in visual control settings. In this work, we study the URLB and propose a new method to solve it, using unsupervised model-based RL, for pre-training the agent, and a task-aware fine-tuning strategy combined with a new proposed hybrid planner, Dyna-MPC, to adapt the agent for downstream tasks. On URLB, our method obtains 93.59% overall normalized performance, surpassing previous baselines by a staggering margin. The approach is empirically evaluated through a large-scale empirical study, which we use to validate our design choices and analyze our models. We also show robust performance on the Real-Word RL benchmark, hinting at resiliency to environment perturbations during adaptation. Project website: https://masteringurlb.github.io/

2021-12-31

arXiv.org (preprint)

Alexia Jolicoeur-Martineau

proceedings.mlr.press

MCVD: Masked Conditional Video Diffusion for Prediction, Generation, and Interpolation

Vikram Voleti

Christopher Pal

2021-12-31

Advances in Neural Information Processing Systems 35 (NeurIPS 2022) (published)

Is a Modular Architecture Enough?

Sarthak Mittal

Yoshua Bengio

Guillaume Lajoie

Inspired from human cognition, machine learning systems are gradually revealing advantages of sparser and more modular architectures. Recent… (see more) work demonstrates that not only do some modular architectures generalize well, but they also lead to better out-of-distribution generalization, scaling properties, learning speed, and interpretability. A key intuition behind the success of such systems is that the data generating system for most real-world settings is considered to consist of sparsely interacting parts, and endowing models with similar inductive biases will be helpful. However, the field has been lacking in a rigorous quantitative assessment of such systems because these real-world data distributions are complex and unknown. In this work, we provide a thorough assessment of common modular architectures, through the lens of simple and known modular data distributions. We highlight the benefits of modularity and sparsity and reveal insights on the challenges faced while optimizing modular systems. In doing so, we propose evaluation metrics that highlight the benefits of modularity, the regimes in which these benefits are substantial, as well as the sub-optimality of current end-to-end learned modular systems as opposed to their claimed potential.

2021-12-31

Advances in Neural Information Processing Systems 35 (NeurIPS 2022) (published)

Multilingual Language Model Adaptive Fine-Tuning: A Study on African Languages

Jesujoba Oluwadara Alabi

David Ifeoluwa Adelani

Marius Mosbach

Dietrich Klakow

and XLM-R) and three NLP tasks (NER, news topic classiﬁcation, and sentiment classiﬁcation) shows that our approach is competitive to ap… (see more)plying LAFT on individual languages while requiring signiﬁcantly less disk space. Finally, we show that our adapted PLM also improves the zero-shot cross-lingual transfer abilities of parameter efﬁcient ﬁne-tuning methods.

2021-12-31

arXiv.org (preprint)

NeoRS: a neonatal resting state fMRI data preprocessing pipeline

V. Enguix

J. Kenley

D. Luck

J. Cohen-Adad

G.A. Lodygensky

Resting state fMRI (rsfMRI) has been shown to be a promising tool to study intrinsic functional connectivity and assess its integrity in cer… (see more)ebral development. In neonates, where fMRI is limited to few paradigms, rsfMRI was shown to be a relevant tool to explore regional interactions of brain networks. However, to identify the resting state networks, data needs to be carefully processed. Because of the non-collaborative nature of the neonates, the differences in brain size and the reversed contrast compared to adults, neonates can't be processed with the existing adult pipelines. Therefore, we developed NeoRS. The main processing steps include atlas registration, skull tripping, segmentation, slice timing and head motion correction and confounds regression. To address the specificity of neonatal brain imaging, particular attention was given to registration including neonatal atlas type and parameters, such as brain size variations, and contrast differences compared to adults. Furthermore, head motion was scrutinized and optimized, as it is a major issue when processing neonatal data. The pipeline includes visual quality control assessment checkpoints. To assess its effectiveness, we used the data from the Baby Connectome Project including 10 neonates. NeoRS was designed to work on both multi-band and single-band acquisitions and is applicable on smaller datasets. It also includes popular functional connectivity analysis features such as seed based correlations. Language, default mode, dorsal attention, visual, ventral attention, motor and fronto parietal networks were evaluated. The different analyzed networks were in agreement with previously published studies in the neonate. NeoRS is coded in Matlab, it is open-source and available on https://github.com/venguix/NeoRS. NeoRS allows robust image processing of the neonatal rsfMRI data that can be readily customized to different datasets.

2021-12-31

Frontiers in Neuroinformatics (published)

Neural Attentive Circuits

Nasim Rahaman

Martin Weiss

Francesco Locatello

Chris Pal

Yoshua Bengio

Bernhard Schölkopf

Li Erran Li

Nicolas Ballas

Recent work has seen the development of general purpose neural architectures that can be trained to perform tasks across diverse data modali… (see more)ties. General purpose models typically make few assumptions about the underlying data-structure and are known to perform well in the large-data regime. At the same time, there has been growing interest in modular neural architectures that represent the data using sparsely interacting modules. These models can be more robust out-of-distribution, computationally efficient, and capable of sample-efficient adaptation to new data. However, they tend to make domain-specific assumptions about the data, and present challenges in how module behavior (i.e., parameterization) and connectivity (i.e., their layout) can be jointly learned. In this work, we introduce a general purpose, yet modular neural architecture called Neural Attentive Circuits (NACs) that jointly learns the parameterization and a sparse connectivity of neural modules without using domain knowledge. NACs are best understood as the combination of two systems that are jointly trained end-to-end: one that determines the module configuration and the other that executes it on an input. We demonstrate qualitatively that NACs learn diverse and meaningful module configurations on the NLVR2 dataset without additional supervision. Quantitatively, we show that by incorporating modularity in this way, NACs improve upon a strong non-modular baseline in terms of low-shot adaptation on CIFAR and CUBs dataset by about 10%, and OOD robustness on Tiny ImageNet-R by about 2.5%. Further, we find that NACs can achieve an 8x speedup at inference time while losing less than 3% performance. Finally, we find NACs to yield competitive results on diverse data modalities spanning point-cloud classification, symbolic processing and text-classification from ASCII bytes, thereby confirming its general purpose nature.

2021-12-31

Advances in Neural Information Processing Systems 35 (NeurIPS 2022) (published)

Optimizing deep learning for Magnetoencephalography (MEG): From sensory perception to sex prediction and brain fingerprinting

Arthur Dehgan

Irina Rish

Karim Jerbi

2021-12-31

2022 Conference on Cognitive Computational Neuroscience (published)