Portrait de Emmanuel Bengio

Emmanuel Bengio

Membre industriel associé
Scientifique en apprentissage automatique, Recursion
Sujets de recherche
Apprentissage par renforcement
Apprentissage profond
GFlowNets
Modèles génératifs
Modélisation moléculaire

Biographie

Emmanuel Bengio est chercheur en ML à Valence Labs/Recursion, où il travaille sur l'intersection des GFlowNets et de la découverte de médicaments. Il a fait son doctorat sous la direction de Joelle Pineau et Doina Precup à McGill/Mila - Institut québécois d'intelligence artificielle, en se concentrant sur la compréhension de la généralisation dans la RL profonde.

Publications

Virtual Cells: Predict, Explain, Discover
Emmanuel Noutahi
Jason Hartford
Prudencio Tossou
Shawn Whitfield
Ali Denton
Cas Wognum
Kristina Ulicna
Jonathan Hsu
Michael Cuccarese
Christopher Gibson
Daniel Cohen
Berton Earnshaw
Virtual Cells: Predict, Explain, Discover
Emmanuel Noutahi
Jason Hartford
Prudencio Tossou
Shawn Whitfield
Ali Denton
Cas Wognum
Kristina Ulicna
Michael Craig
Jonathan Hsu
Michael Cuccarese
Christopher Gibson
Daniel Cohen
Berton Earnshaw
Solving Bayesian inverse problems with diffusion priors and off-policy RL
Luca Scimeca
Siddarth Venkatraman
Moksh J. Jain
Minsu Kim
Marcin Sendera
Mohsin Hasan
Luke Rowe
Sarthak Mittal
Pablo Lemos
Alexandre Adam
Jarrid Rector-Brooks
Nikolay Malkin
This paper presents a practical application of Relative Trajectory Balance (RTB), a recently introduced off-policy reinforcement learning (R… (voir plus)L) objective that can asymptotically solve Bayesian inverse problems optimally. We extend the original work by using RTB to train conditional diffusion model posteriors from pretrained unconditional priors for challenging linear and non-linear inverse problems in vision, and science. We use the objective alongside techniques such as off-policy backtracking exploration to improve training. Importantly, our results show that existing training-free diffusion posterior methods struggle to perform effective posterior inference in latent space due to inherent biases.
Pretraining Generative Flow Networks with Inexpensive Rewards for Molecular Graph Generation
Mohit Pandey
Gopeshh Subbaraj
Artem Cherkasov
Martin Ester
Action abstractions for amortized sampling
Oussama Boussif
Lena Nehale Ezzine
Joseph D Viviano
Michał Koziarski
Moksh J. Jain
Nikolay Malkin
Rim Assouel
Adaptive teachers for amortized samplers
Minsu Kim
Sanghyeok Choi
Taeyoung Yun
Leo Feng
Jarrid Rector-Brooks
Sungsoo Ahn
Jinkyoo Park
Nikolay Malkin
Amortized inference is the task of training a parametric model, such as a neural network, to approximate a distribution with a given unnorma… (voir plus)lized density where exact sampling is intractable. When sampling is implemented as a sequential decision-making process, reinforcement learning (RL) methods, such as generative flow networks, can be used to train the sampling policy. Off-policy RL training facilitates the discovery of diverse, high-reward candidates, but existing methods still face challenges in efficient exploration. We propose to use an adaptive training distribution (the Teacher) to guide the training of the primary amortized sampler (the Student) by prioritizing high-loss regions. The Teacher, an auxiliary behavior model, is trained to sample high-error regions of the Student and can generalize across unexplored modes, thereby enhancing mode coverage by providing an efficient training curriculum. We validate the effectiveness of this approach in a synthetic environment designed to present an exploration challenge, two diffusion-based sampling tasks, and four biochemical discovery tasks demonstrating its ability to improve sample efficiency and mode coverage.
Towards Improving Exploration through Sibling Augmented GFlowNets
Kanika Madan
Alex Lamb
Exploration is a key factor for the success of an active learning agent, especially when dealing with sparse extrinsic terminal rewards and … (voir plus)long trajectories. We introduce Sibling Augmented Generative Flow Networks (SA-GFN), a novel framework designed to enhance exploration and training efficiency of Generative Flow Networks (GFlowNets). SA-GFN uses a decoupled dual network architecture, comprising of a main Behavior Network and an exploratory Sibling Network, to enable a diverse exploration of the underlying distribution using intrinsic rewards. Inspired by the ideas on exploration from reinforcement learning, SA-GFN provides a general-purpose exploration and learning paradigm that integrates with multiple GFlowNet training objectives and is especially helpful for exploration over a wide range of sparse or low reward distributions and task structures. An extensive set of experiments across a diverse range of tasks, reward structures and trajectory lengths, along with a thorough set of ablations, demonstrate the superior performance of SA-GFN in terms of exploration efficacy and convergence speed as compared to the existing methods. In addition, SA-GFN's versatility and compatibility with different GFlowNet training objectives and intrinsic reward methods underscores its broad applicability in various problem domains.
SynFlowNet: Design of Diverse and Novel Molecules with Synthesis Constraints
M. Cretu
Charles Harris
Ilia Igashov
Arne Schneuing
Marwin Segler
Bruno Correia
Julien Roy
Pietro Lio
Generative models see increasing use in computer-aided drug design. However, while performing well at capturing distributions of molecular m… (voir plus)otifs, they often produce synthetically inaccessible molecules. To address this, we introduce SynFlowNet, a GFlowNet model whose action space uses chemical reactions and buyable reactants to sequentially build new molecules. By incorporating forward synthesis as an explicit constraint of the generative mechanism, we aim at bridging the gap between in silico molecular generation and real world synthesis capabilities. We evaluate our approach using synthetic accessibility scores and an independent retrosynthesis tool to assess the synthesizability of our compounds, and motivate the choice of GFlowNets through considerable improvement in sample diversity compared to baselines. Additionally, we identify challenges with reaction encodings that can complicate traversal of the MDP in the backward direction. To address this, we introduce various strategies for learning the GFlowNet backward policy and thus demonstrate how additional constraints can be integrated into the GFlowNet MDP framework. This approach enables our model to successfully identify synthesis pathways for previously unseen molecules.
Efficient Biological Data Acquisition through Inference Set Design
Ihor Neporozhnii
Julien Roy
Jason Hartford
In drug discovery, highly automated high-throughput laboratories are used to screen a large number of compounds in search of effective drugs… (voir plus). These experiments are expensive, so one might hope to reduce their cost by only experimenting on a subset of the compounds, and predicting the outcomes of the remaining experiments. In this work, we model this scenario as a sequential subset selection problem: we aim to select the smallest set of candidates in order to achieve some desired level of accuracy for the system as a whole. Our key observation is that, if there is heterogeneity in the difficulty of the prediction problem across the input space, selectively obtaining the labels for the hardest examples in the acquisition pool will leave only the relatively easy examples to remain in the inference set, leading to better overall system performance. We call this mechanism inference set design, and propose the use of a confidence-based active learning solution to prune out these challenging examples. Our algorithm includes an explicit stopping criterion that interrupts the acquisition loop when it is sufficiently confident that the system has reached the target performance. Our empirical studies on image and molecular datasets, as well as a real-world large-scale biological assay, show that active learning for inference set design leads to significant reduction in experimental cost while retaining high system performance.
Efficient Biological Data Acquisition through Inference Set Design
Ihor Neporozhnii
Julien Roy
Jason Hartford
In drug discovery, highly automated high-throughput laboratories are used to screen a large number of compounds in search of effective drugs… (voir plus). These experiments are expensive, so one might hope to reduce their cost by only experimenting on a subset of the compounds, and predicting the outcomes of the remaining experiments. In this work, we model this scenario as a sequential subset selection problem: we aim to select the smallest set of candidates in order to achieve some desired level of accuracy for the system as a whole. Our key observation is that, if there is heterogeneity in the difficulty of the prediction problem across the input space, selectively obtaining the labels for the hardest examples in the acquisition pool will leave only the relatively easy examples to remain in the inference set, leading to better overall system performance. We call this mechanism inference set design, and propose the use of a confidence-based active learning solution to prune out these challenging examples. Our algorithm includes an explicit stopping criterion that interrupts the acquisition loop when it is sufficiently confident that the system has reached the target performance. Our empirical studies on image and molecular datasets, as well as a real-world large-scale biological assay, show that active learning for inference set design leads to significant reduction in experimental cost while retaining high system performance.
Improved Off-policy Reinforcement Learning in Biological Sequence Design
Hyeonah Kim
Minsu Kim
Taeyoung Yun
Sanghyeok Choi
Alex Hern'andez-Garc'ia
Jinkyoo Park
Designing biological sequences with desired properties is a significant challenge due to the combinatorially vast search space and the high … (voir plus)cost of evaluating each candidate sequence. To address these challenges, reinforcement learning (RL) methods, such as GFlowNets, utilize proxy models for rapid reward evaluation and annotated data for policy training. Although these approaches have shown promise in generating diverse and novel sequences, the limited training data relative to the vast search space often leads to the misspecification of proxy for out-of-distribution inputs. We introduce
Improved Off-policy Reinforcement Learning in Biological Sequence Design
Hyeonah Kim
Minsu Kim
Taeyoung Yun
Sanghyeok Choi
Alex Hernandez-Garcia
Jinkyoo Park
Designing biological sequences with desired properties is a significant challenge due to the combinatorially vast search space and the high … (voir plus)cost of evaluating each candidate sequence. To address these challenges, reinforcement learning (RL) methods, such as GFlowNets, utilize proxy models for rapid reward evaluation and annotated data for policy training. Although these approaches have shown promise in generating diverse and novel sequences, the limited training data relative to the vast search space often leads to the misspecification of proxy for out-of-distribution inputs. We introduce