Publications

Goal-conditioned GFlowNets for Controllable Multi-Objective Molecular Design
Julien Roy
Pierre-Luc Bacon
Christopher Pal
Emmanuel Bengio
In recent years, in-silico molecular design has received much attention from the machine learning community. When designing a new compound f… (voir plus)or pharmaceutical applications, there are usually multiple properties of such molecules that need to be optimised: binding energy to the target, synthesizability, toxicity, EC50, and so on. While previous approaches have employed a scalarization scheme to turn the multi-objective problem into a preference-conditioned single objective, it has been established that this kind of reduction may produce solutions that tend to slide towards the extreme points of the objective space when presented with a problem that exhibits a concave Pareto front. In this work we experiment with an alternative formulation of goal-conditioned molecular generation to obtain a more controllable conditional model that can uniformly explore solutions along the entire Pareto front.
2023-06-22
ICML.cc/2023/Workshop/DeployableGenerativeAI (accepté)
doi.org
openreview.net
Using modular connectome-based predictive modeling to reveal brain-behavior relationships of individual differences in working memory
Huayi Yang
Junjun Zhang
Zhenlan Jin
Pouya Bashivan
Ling Li
2023-06-21
Brain Structure and Function (publié)
doi.org
Accelerating exploration and representation learning with offline pre-training
Bogdan Mazoure
Jacob Bruce
Doina Precup
Rob Fergus
Ankit Anand
Sequential decision-making agents struggle with long horizon tasks, since solving them requires multi-step reasoning. Most reinforcement lea… (voir plus)rning (RL) algorithms address this challenge by improved credit assignment, introducing memory capability, altering the agent's intrinsic motivation (i.e. exploration) or its worldview (i.e. knowledge representation). Many of these components could be learned from offline data. In this work, we follow the hypothesis that exploration and representation learning can be improved by separately learning two different models from a single offline dataset. We show that learning a state representation using noise-contrastive estimation and a model of auxiliary reward separately from a single collection of human demonstrations can significantly improve the sample efficiency on the challenging NetHack benchmark. We also ablate various components of our experimental setting and highlight crucial insights.
2023-06-19
ICML.cc/2023/Workshop/ILHF (accepté)
doi.org
openreview.net
Cognitive Models as Simulators: Using Cognitive Models to Tap into Implicit Human Feedback
Ardavan S. Nobandegani
Thomas Shultz
Irina Rish
2023-06-19
ICML.cc/2023/Workshop/ILHF (accepté)
openreview.net
openreview.net
Constant Memory Attention Block
Leo Feng
Frederick Tung
Hossein Hajimirsadeghi
Yoshua Bengio
Mohamed Osama Ahmed
2023-06-19
ICML.cc/2023/Workshop/ES-FoMO (poster)
doi.org
openreview.net
Continual Pre-Training of Large Language Models: How to (re)warm your model?
Kshitij Gupta
Benjamin Therien
Adam Ibrahim
Mats Leon Richter
Quentin Gregory Anthony
Eugene Belilovsky
Irina Rish
Timothee LESORT
Large language models (LLMs) are routinely pre-trained on billions of tokens, only to restart the process over again once new data becomes a… (voir plus)vailable. A much cheaper and more efficient solution would be to enable the continual pre-training of these models, i.e. updating pre-trained models with new data instead of re-training them from scratch. However, the distribution shift induced by novel data typically results in degraded performance on past data. Taking a step towards efficient continual pre-training, in this work, we examine the effect of different warm-up strategies. Our hypothesis is that the learning rate must be re-increased to improve compute efficiency when training on a new dataset. We study the warmup phase of models pre-trained on the Pile (upstream data, 300B tokens) as we continue to pre-train on SlimPajama (downstream data, 297B tokens), following a linear warmup and cosine decay schedule. We conduct all experiments on the Pythia 410M language model architecture and evaluate performance through validation perplexity. We experiment with different pre-training checkpoints, various maximum learning rates, and various warmup lengths. Our results show that while rewarming models first increases the loss on upstream and downstream data, in the longer run it improves the downstream performance, outperforming models trained from scratch
2023-06-19
ES-FoMO @ International Conference on Machine Learning (poster)
doi.org
openreview.net
Deep Clustering with Self-supervision using Pairwise Data Similarities
Mohammadreza Sadeghi
Narges Armanfard
Deep clustering incorporates embedding into clustering to find a lower-dimensional space appropriate for clustering. Most of the existing me… (voir plus)thods try to group similar data points through simultaneously minimizing clustering and reconstruction losses, employing an autoencoder (AE). However, they all ignore the relevant useful information available within pairwise data relationships. In this paper we propose a novel deep clustering framework with self-supervision using pairwise data similarities (DCSS). The proposed method consists of two successive phases. First, we propose a novel AE-based approach that aims to aggregate similar data points near a common group center in the latent space of an AE. The AE's latent space is obtained by minimizing weighted reconstruction and centering losses of data points, where weights are defined based on similarity of data points and group centers. In the second phase, we map the AE's latent space, using a fully connected network MNet, onto a K-dimensional space used to derive the final data cluster assignments, where K is the number of clusters. MNet is trained to strengthen (weaken) similarity of similar (dissimilar) samples. Experimental results on multiple benchmark datasets demonstrate the effectiveness of DCSS for data clustering and as a general framework for boosting up state-of-the-art clustering methods.
2023-06-19
INDIGO (prépublication)
doi.org
GFlowNets for Causal Discovery: an Overview
Dragos Cristian Manta
Edward J Hu
Yoshua Bengio
2023-06-19
ICML.cc/2023/Workshop/SODS (poster)
openreview.net
openreview.net
ROSA: Random Orthogonal Subspace Adaptation
Marawan Gamal
Guillaume Rabusseau
Aristides Milios
Siva Reddy
2023-06-19
ICML.cc/2023/Workshop/ES-FoMO (poster)
openreview.net
openreview.net
Self-Supervised Image-to-Point Distillation via Semantically Tolerant Contrastive Loss
Anas Mahmoud
Jordan S. K. Hu
Tianshu Kuai
Ali Harakeh
Liam Paull
Steven L. Waslander
An effective framework for learning 3D representations for perception tasks is distilling rich self-supervised image features via contrastiv… (voir plus)e learning. However, image-to point representation learning for autonomous driving datasets faces two main challenges: 1) the abundance of self-similarity, which results in the contrastive losses pushing away semantically similar point and image regions and thus disturbing the local semantic structure of the learned representations, and 2) severe class imbalance as pretraining gets dominated by over-represented classes. We propose to alleviate the self-similarity problem through a novel semantically tolerant image-to-point contrastive loss that takes into consideration the semantic distance between positive and negative image regions to minimize contrasting semantically similar point and image regions. Additionally, we address class imbalance by designing a class-agnostic balanced loss that approximates the degree of class imbalance through an aggregate sample-to-samples semantic similarity measure. We demonstrate that our semantically-tolerant contrastive loss with class balancing improves state-of-the art 2D-to-3D representation learning in all evaluation settings on 3D semantic segmentation. Our method consistently outperforms state-of-the-art 2D-to-3D representation learning frameworks across a wide range of 2D self-supervised pretrained models.
2023-06-19
IEEE/CVF Conference on Computer Vision and Pattern Recognition (Accept)
doi.org
arxiv.org
Simulated Annealing in Early Layers Leads to Better Generalization
Amir M. Sarfi
Zahra Karimpour
Muawiz Chaudhary
Nasir M. Khalid
Mirco Ravanelli
Sudhir Mudur
Eugene Belilovsky
Recently, a number of iterative learning methods have been introduced to improve generalization. These typically rely on training for longer… (voir plus) periods of time in exchange for improved generalization. LLF (later-layer-forgetting) is a state-of-the-art method in this category. It strengthens learning in early layers by periodically re-initializing the last few layers of the network. Our principal innovation in this work is to use Simulated annealing in EArly Layers (SEAL) of the network in place of re-initialization of later layers. Essentially, later layers go through the normal gradient descent process, while the early layers go through short stints of gradient ascent followed by gradient descent. Extensive experiments on the popular Tiny-ImageNet dataset benchmark and a series of transfer learning and few-shot learning tasks show that we outperform LLF by a significant margin. We further show that, compared to normal training, LLF features, although improving on the target task, degrade the transfer learning performance across all datasets we explored. In comparison, our method outperforms LLF across the same target datasets by a large margin. We also show that the prediction depth of our method is significantly lower than that of LLF and normal training, indicating on average better prediction performance.
2023-06-19
IEEE/CVF Conference on Computer Vision and Pattern Recognition (Accept)
doi.org
arxiv.org
Towards Out-of-Distribution Adversarial Robustness
Adam Ibrahim
Charles Guille-escuret
Ioannis Mitliagkas
Irina Rish
David Krueger
Pouya Bashivan
Adversarial robustness continues to be a major challenge for deep learning. A core issue is that robustness to one type of attack often fail… (voir plus)s to transfer to other attacks. While prior work establishes a theoretical trade-off in robustness against different
2023-06-19
ICML.cc/2023/Workshop/AdvML-Frontiers (publié)
doi.org
openreview.net