Portrait of Nicolas Le Roux

Nicolas Le Roux

Core Industry Member
Canada CIFAR AI Chair
Adjunct Professor, McGill University, School of Computer Science
Adjunct Professor, Université de Montréal, Department of Computer Science and Operations Research
Research Scientist, Microsoft Research
Research Topics
Deep Learning
Generative Models
Optimization
Reinforcement Learning

Biography

I am an academic researcher with expertise in machine learning, computer vision, neural networks, deep learning, optimization, large-scale learning and statistical modelling in general.

Current Students

Arnaud Bergeron
Master's Research - Université de Montréal
Alan Chan
PhD - Université de Montréal
Co-supervisor :
David Scott Krueger
Website
Google Scholar
Jacob Chmura
Master's Research - McGill University
Co-supervisor :
Reihaneh Rabbany
Website
Github
Ekaterina Lobacheva
Postdoctorate - Université de Montréal
Co-supervisor :
Irina Rish
Website
Github
Google Scholar

Publications

Tapered Off-Policy REINFORCE: Stable and efficient reinforcement learning for LLMs
Nicolas Le Roux
Marc Gendron-Bellemare
Jonathan Lebensold
Arnaud Bergeron
Joshua Greaves
Alex Fr'echette
Carolyne Pelletier
Eric Thibodeau-Laufer
S'andor Toth
Sam Work
2025-03-18
ArXiv (preprint)
arxiv.org
arxiv.org
Fast Convergence of Softmax Policy Mirror Ascent
Reza Asad
Reza Babanezhad Harikandeh
Issam Hadj Laradji
Nicolas Le Roux
Sharan Vaswani
Natural policy gradient (NPG) is a common policy optimization algorithm and can be viewed as mirror ascent in the space of probabilities. Re… (see more)cently,~\citet{vaswani2021general} introduced a policy gradient method that corresponds to mirror ascent in the dual space of logits. We refine this algorithm, removing its need for a normalization across actions and analyze the resulting method (referred to as SPMA). For tabular MDPs, we prove that SPMA with a constant step-size matches the linear convergence of NPG and achieves a faster convergence than constant step-size (accelerated) softmax policy gradient. To handle large state-action spaces, we extend SPMA to use a log-linear policy parameterization. Unlike that for NPG, generalizing SPMA to the linear function approximation (FA) setting does not require compatible function approximation. Unlike MDPO, a practical generalization of NPG, SPMA with linear FA only requires solving convex softmax classification problems. We prove that SPMA achieves linear convergence to the neighbourhood of the optimal value function. We extend SPMA to handle non-linear FA and evaluate its empirical performance on the MuJoCo and Atari benchmarks. Our results demonstrate that SPMA consistently achieves similar or better performance compared to MDPO, PPO and TRPO.
2025-01-22
aistats.org/AISTATS/2025/Conference (poster)
openreview.net
openreview.net
Fast Convergence of Softmax Policy Mirror Ascent for Bandits & Tabular MDPs
Reza Asad
Reza Babanezhad Harikandeh
Issam Hadj Laradji
Nicolas Le Roux
Sharan Vaswani
We analyze the convergence of a novel policy gradient algorithm (referred to as SPMA) for multi-armed bandits and tabular Markov decision pr… (see more)ocesses (MDPs). SPMA is an instantiation of mirror ascent and uses the softmax parameterization with a log-sum-exp mirror map. Given access to the exact policy gradients, we prove that SPMA with a constant step-size requires
2024-10-10
NeurIPS.cc/2024/Workshop/OPT (published)
openreview.net
openreview.net
How Learning Rates Shape Neural Network Focus: Insights from Example Ranking
Ekaterina Lobacheva
Keller Jordan
Aristide Baratin
Nicolas Le Roux
The learning rate is a key hyperparameter that affects both the speed of training and the generalization performance of neural networks. Th… (see more)rough a new {\it loss-based example ranking} analysis, we show that networks trained with different learning rates focus their capacity on different parts of the data distribution, leading to solutions with different generalization properties. These findings, which hold across architectures and datasets, provide new insights into how learning rates affect model performance and example-level dynamics in neural networks.
2024-10-10
NeurIPS.cc/2024/Workshop/SciForDL (poster)
openreview.net
openreview.net
VinePPO: Accurate Credit Assignment in RL for LLM Mathematical Reasoning
Amirhossein Kazemnejad
Milad Aghajohari
Eva Portelance
Alessandro Sordoni
Siva Reddy
Aaron Courville
Nicolas Le Roux
Large language models (LLMs) are increasingly required to solve complex reasoning tasks, like mathematical problems, that involve multiple r… (see more)easoning steps before feedback is received. Effectively identifying and prioritizing key steps by accurately assigning credit to these intermediate steps is essential for enhancing model performance. Proximal Policy Optimization (PPO), a state-of-the-art reinforcement learning algorithm for finetuning LLMs, addresses the credit assignment problem by employing value networks to predict the expected cumulative rewards of intermediate states. In this work, we identify significant limitations with this value estimation method. To address this, we propose \methodname that leverages the flexibility of language environments to compute unbiased Monte Carlo-based estimates of the intermediate values. VinePPO consistently outperforms standard PPO, doing so more efficiently and with lower divergence from the reference model. Our findings underscore the critical importance of accurate credit assignment in LLM post-training and present a simple, yet effective solution.
2024-10-09
NeurIPS.cc/2024/Workshop/MATH-AI (accepted)
openreview.net
openreview.net
VinePPO: Accurate Credit Assignment in RL for LLM Mathematical Reasoning
Amirhossein Kazemnejad
Milad Aghajohari
Eva Portelance
Alessandro Sordoni
Siva Reddy
Aaron Courville
Nicolas Le Roux
Large language models (LLMs) are increasingly required to solve complex reasoning tasks, like mathematical problems, that involve multiple r… (see more)easoning steps before feedback is received. Effectively identifying and prioritizing key steps by accurately assigning credit to these intermediate steps is essential for enhancing model performance. Proximal Policy Optimization (PPO), a state-of-the-art reinforcement learning algorithm for finetuning LLMs, addresses the credit assignment problem by employing value networks to predict the expected cumulative rewards of intermediate states. In this work, we identify significant limitations with this value estimation method. To address this, we propose \methodname that leverages the flexibility of language environments to compute unbiased Monte Carlo-based estimates of the intermediate values. VinePPO consistently outperforms standard PPO, doing so more efficiently and with lower divergence from the reference model. Our findings underscore the critical importance of accurate credit assignment in LLM post-training and present a simple, yet effective solution.
2024-10-09
NeurIPS.cc/2024/Workshop/MATH-AI (accepted)
openreview.net
openreview.net
fPLSA: Learning Semantic Structures in Document Collections Using Foundation Models
Weijia Xu
Nebojsa Jojic
Nicolas Le Roux
2024-10-07
ArXiv (preprint)
doi.org
openreview.net
VinePPO: Unlocking RL Potential For LLM Reasoning Through Refined Credit Assignment
Amirhossein Kazemnejad
Milad Aghajohari
Eva Portelance
Alessandro Sordoni
Siva Reddy
Aaron Courville
Nicolas Le Roux
2024-10-02
ArXiv (preprint)
doi.org
openreview.net
VinePPO: Unlocking RL Potential For LLM Reasoning Through Refined Credit Assignment
Amirhossein Kazemnejad
Milad Aghajohari
Eva Portelance
Alessandro Sordoni
Siva Reddy
Aaron Courville
Nicolas Le Roux
Large language models (LLMs) are increasingly applied to complex reasoning tasks that require executing several complex steps before receivi… (see more)ng any reward. Properly assigning credit to these steps is essential for enhancing model performance. Proximal Policy Optimization (PPO), a state-of-the-art reinforcement learning (RL) algorithm used for LLM finetuning, employs value networks to tackle credit assignment. However, value networks face challenges in predicting the expected cumulative rewards accurately in complex reasoning tasks, often leading to high-variance updates and suboptimal performance. In this work, we systematically evaluate the efficacy of value networks and reveal their significant shortcomings in reasoning-heavy LLM tasks, showing that they barely outperform a random baseline when comparing alternative steps. To address this, we propose VinePPO, a straightforward approach that leverages the flexibility of language environments to compute unbiased Monte Carlo-based estimates, bypassing the need for large value networks. Our method consistently outperforms PPO and other RL-free baselines across MATH and GSM8K datasets with fewer gradient updates (up to 9x), less wall-clock time (up to 3.0x). These results emphasize the importance of accurate credit assignment in RL finetuning of LLM and demonstrate VinePPO's potential as a superior alternative.
2024-10-02
ArXiv (preprint)
doi.org
arxiv.org
VinePPO: Unlocking RL Potential For LLM Reasoning Through Refined Credit Assignment
Amirhossein Kazemnejad
Milad Aghajohari
Eva Portelance
Alessandro Sordoni
Siva Reddy
Aaron Courville
Nicolas Le Roux
Large language models (LLMs) are increasingly applied to complex reasoning tasks that require executing several complex steps before receivi… (see more)ng any reward. Properly assigning credit to these steps is essential for enhancing model performance. Proximal Policy Optimization (PPO), a state-of-the-art reinforcement learning (RL) algorithm used for LLM finetuning, employs value networks to tackle credit assignment. However, value networks face challenges in predicting the expected cumulative rewards accurately in complex reasoning tasks, often leading to high-variance updates and suboptimal performance. In this work, we systematically evaluate the efficacy of value networks and reveal their significant shortcomings in reasoning-heavy LLM tasks, showing that they barely outperform a random baseline when comparing alternative steps. To address this, we propose VinePPO, a straightforward approach that leverages the flexibility of language environments to compute unbiased Monte Carlo-based estimates, bypassing the need for large value networks. Our method consistently outperforms PPO and other RL-free baselines across MATH and GSM8K datasets with fewer gradient updates (up to 9x), less wall-clock time (up to 3.0x). These results emphasize the importance of accurate credit assignment in RL finetuning of LLM and demonstrate VinePPO's potential as a superior alternative.
2024-10-02
ArXiv (preprint)
doi.org
arxiv.org
Improving Context-Aware Preference Modeling for Language Models
Silviu Pitis
Ziang Xiao
Nicolas Le Roux
Alessandro Sordoni
While finetuning language models from pairwise preferences has proven remarkably effective, the underspecified nature of natural language pr… (see more)esents critical challenges. Direct preference feedback is uninterpretable, difficult to provide where multidimensional criteria may apply, and often inconsistent, either because it is based on incomplete instructions or provided by diverse principals. To address these challenges, we consider the two-step preference modeling procedure that first resolves the under-specification by selecting a context, and then evaluates preference with respect to the chosen context. We decompose reward modeling error according to these two steps, which suggests that supervising context in addition to context-specific preference may be a viable approach to aligning models with diverse human preferences. For this to work, the ability of models to evaluate context-specific preference is critical. To this end, we contribute context-conditioned preference datasets and accompanying experiments that investigate the ability of language models to evaluate context-specific preference. Unlike past datasets, where context-specific preference is highly correlated with general preference, our "preference reversal" datasets disentangle context-specific and general preferences to isolate context-specific capabilities. We use our datasets to (1) show that existing preference models benefit from, but fail to fully consider, added context, (2) finetune a context-aware reward model with context-specific performance exceeding that of GPT-4 and Llama 3 70B, and (3) investigate the potential value of context-aware preference modeling.
2024-09-25
NeurIPS.cc/2024/Conference (poster)
doi.org
openreview.net
Improving Context-Aware Preference Modeling for Language Models
Silviu Pitis
Ziang Xiao
Nicolas Le Roux
Alessandro Sordoni
While finetuning language models from pairwise preferences has proven remarkably effective, the underspecified nature of natural language pr… (see more)esents critical challenges. Direct preference feedback is uninterpretable, difficult to provide where multidimensional criteria may apply, and often inconsistent, either because it is based on incomplete instructions or provided by diverse principals. To address these challenges, we consider the two-step preference modeling procedure that first resolves the under-specification by selecting a context, and then evaluates preference with respect to the chosen context. We decompose reward modeling error according to these two steps, which suggests that supervising context in addition to context-specific preference may be a viable approach to aligning models with diverse human preferences. For this to work, the ability of models to evaluate context-specific preference is critical. To this end, we contribute context-conditioned preference datasets and accompanying experiments that investigate the ability of language models to evaluate context-specific preference. We use our datasets to (1) show that existing preference models benefit from, but fail to fully consider, added context, (2) finetune a context-aware reward model with context-specific performance exceeding that of GPT-4 and Llama 3 70B on tested datasets, and (3) investigate the value of context-aware preference modeling.
2024-07-20
ArXiv (preprint)
doi.org
arxiv.org