Aaron Courville

sarvjeet-singh.ghotra@mila.quebec

Juan Duque

PhD - Université de Montréal

juan.duque@mila.quebec

PhD - Université de Montréal

udaykaran.kapur@mila.quebec

Arian Hosseini

PhD - Université de Montréal

Uday Kapur

PhD - Université de Montréal

Amr Khalifa

PhD - Université de Montréal

amr-khalifa.alshaykh@mila.quebec

Zhixuan Lin

PhD - Université de Montréal

PhD - Université de Montréal

Co-supervisor :

alan.milligan@mila.quebec

waltermayor@gmail.com

PhD - Université de Montréal

Principal supervisor :

Simon Lacoste-Julien

PhD - Université de Montréal

Principal supervisor :

Gauthier Gidel

andjela.mladenovic@mila.quebec

mohammed.muqeeth@mila.quebec

Muqeeth Mohammed

PhD - Université de Montréal

PhD - Université de Montréal

Co-supervisor :

Rishabh Agarwal

morgane.moss@mila.quebec

Andrei Nicolicioiu

PhD - Université de Montréal

andrei.nicolicioiu@gmail.com

Michael Noukhovitch

PhD - Université de Montréal

Johan Samir Obando Ceron

PhD - Université de Montréal

Co-supervisor :

michell.payanoperez@mila.quebec

johan.ceron@mila.quebec

PhD - Université de Montréal

Dereck Piché

PhD - Université de Montréal

pichedereck@gmail.com

khaled.rouissi@mila.quebec

Khaled Rouissi

Master's Research - Université de Montréal

Esra'a Saleh

PhD - Université de Montréal

Principal supervisor :

Glen Berseth

esraa.saleh@mila.quebec

Vedant Shah

PhD - Université de Montréal

vedant.shah@mila.quebec

colin.yiyang-lu@mila.quebec

Yusong Wu

PhD - Université de Montréal

Principal supervisor :

Anna (Cheng-Zhi) Huang

wu.yusong@mila.quebec

Colin Yiyang Lu

PhD - Université de Montréal

Sujin yun

PhD - Université de Montréal

sujin.yun@mila.quebec

xiaofeng.zhang@mila.quebec

Xiaofeng Zhang

PhD - Université de Montréal

Publications

Don't Flatten, Tokenize! Unlocking the Key to SoftMoE's Efficacy in Deep RL

Ghada Sokar

Johan Obando-Ceron

Hugo Larochelle

The use of deep neural networks in reinforcement learning (RL) often suffers from performance degradation as model size increases. While sof… (see more)t mixtures of experts (SoftMoEs) have recently shown promise in mitigating this issue for online RL, the reasons behind their effectiveness remain largely unknown. In this work we provide an in-depth analysis identifying the key factors driving this performance gain. We discover the surprising result that tokenizing the encoder output, rather than the use of multiple experts, is what is behind the efficacy of SoftMoEs. Indeed, we demonstrate that even with an appropriately scaled single expert, we are able to maintain the performance gains, largely thanks to tokenization.

2025-04-22

International Conference on Learning Representations (Accept (Spotlight))

Neuroplastic Expansion in Deep Reinforcement Learning

Jiashun Liu

Johan Obando-Ceron

Ling Pan

The loss of plasticity in learning agents, analogous to the solidification of neural pathways in biological brains, significantly impedes le… (see more)arning and adaptation in reinforcement learning due to its non-stationary nature. To address this fundamental challenge, we propose a novel approach, {\it Neuroplastic Expansion} (NE), inspired by cortical expansion in cognitive science. NE maintains learnability and adaptability throughout the entire training process by dynamically growing the network from a smaller initial size to its full dimension. Our method is designed with three key components: (\textit{1}) elastic topology generation based on potential gradients, (\textit{2}) dormant neuron pruning to optimize network expressivity, and (\textit{3}) neuron consolidation via experience review to strike a balance in the plasticity-stability dilemma. Extensive experiments demonstrate that NE effectively mitigates plasticity loss and outperforms state-of-the-art methods across various tasks in MuJoCo and DeepMind Control Suite environments. NE enables more adaptive learning in complex, dynamic environments, which represents a crucial step towards transitioning deep reinforcement learning from static, one-time training paradigms to more flexible, continually adapting models.

2025-04-22

International Conference on Learning Representations (Accept (Poster))

Scaling Stick-Breaking Attention: An Efficient Implementation and In-Depth Study

Shawn Tan

Yikang Shen

Songlin Yang

Rameswar Panda

The self-attention mechanism traditionally relies on the softmax operator, necessitating positional embeddings like RoPE, or position biases… (see more) to account for token order. But current methods using still face length generalisation challenges. We investigate an alternative attention mechanism based on the stick-breaking process in larger scale settings. The method works as follows: For each token before the current, we determine a break point, which represents the proportion of the stick, the weight of the attention, to allocate to the current token. We repeat this on the remaining stick, until all tokens are allocated a weight, resulting in a sequence of attention weights. This process naturally incorporates recency bias, which has linguistic motivations for grammar parsing. We study the implications of replacing the conventional softmax-based attention mechanism with stick-breaking attention. We then discuss implementation of numerically stable stick-breaking attention and adapt Flash Attention to accommodate this mechanism. When used as a drop-in replacement for current softmax+RoPE attention systems, we find that stick-breaking attention performs competitively with current methods on length generalisation and downstream tasks. Stick-breaking also performs well at length generalisation, allowing a model trained with

2025-04-22

International Conference on Learning Representations (Accept (Poster))

Advantage Alignment Algorithms

The growing presence of artificially intelligent agents in everyday decision-making, from LLM assistants to autonomous vehicles, hints at a … (see more)future in which conflicts may arise from each agent optimizing individual interests. In general-sum games these conflicts are apparent, where naive Reinforcement Learning agents get stuck in Pareto-suboptimal Nash equilibria. Consequently, opponent shaping has been introduced as a method with success at finding socially beneficial equilibria in social dilemmas. In this work, we introduce Advantage Alignment, a family of algorithms derived from first principles that perform opponent shaping efficiently and intuitively. This is achieved by aligning the advantages of conflicting agents in a given game by increasing the probability of mutually-benefiting actions. We prove that existing opponent shaping methods, including LOLA and LOQA, implicitly perform Advantage Alignment. Compared to these works, Advantage Alignment mathematically simplifies the formulation of opponent shaping and seamlessly works for continuous action domains. We also demonstrate the effectiveness of our algorithm in a wide range of social dilemmas, achieving state of the art results in each case, including a social dilemma version of the Negotiation Game.

2025-01-21

International Conference on Learning Representations (oral)

Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models

Shengyi Huang

The dominant paradigm for RLHF is online and on-policy RL: synchronously generating from the large language model (LLM) policy, labelling wi… (see more)th a reward model, and learning using feedback on the LLM's own outputs. While performant, this paradigm is computationally inefficient. Inspired by classical deep RL literature, we propose separating generation and learning in RLHF. This enables asynchronous generation of new samples while simultaneously training on old samples, leading to faster training and more compute-optimal scaling. However, asynchronous training relies on an underexplored regime, online but off-policy RLHF: learning on samples from previous iterations of our model which give a worse training signal. We tackle the fundamental challenge in this regime: how much off-policyness can we tolerate for asynchronous training to speed up learning but maintain performance? Among several RLHF algorithms we test, online DPO is found to be most robust to off-policy data, and robustness increases with the scale of the policy model. We study further compute optimizations for asynchronous RLHF but find that they come at a performance cost, giving rise to a trade-off. We verify the scalability of asynchronous RLHF by training a general-purpose chatbot from LLaMA 3.1 8B on an instruction-following task ~40% faster than a synchronous run while matching final performance. Finally, we extend our results to math and reasoning to demonstrate asynchronous RL can finetune Rho 1B on GSM8k ~70% faster while matching synchronous accuracy.

2025-01-21

International Conference on Learning Representations (poster)

Forgetting Transformer: Softmax Attention with a Forget Gate

Zhixuan Lin

Evgenii Nikishin

Xu He

An essential component of modern recurrent sequence models is the forget gate. While Transformers do not have an explicit recurrent form, we… (see more) show that a forget gate can be naturally incorporated into Transformers by down-weighting the unnormalized attention scores in a data-dependent way. We name this attention mechanism Forgetting Attention and the resulting model the Forgetting Transformer (FoX). We show that FoX outperforms the Transformer on long-context language modeling, length extrapolation, and short-context downstream tasks, while performing on par with the Transformer on long-context downstream tasks. Moreover, it is compatible with the FlashAttention algorithm and does not require any positional embeddings. Several analyses, including the needle-in-the-haystack test, show that FoX also retains the Transformer's superior long-context capabilities over recurrent sequence models such as Mamba-2, HGRN2, and DeltaNet. We also introduce a "Pro" block design that incorporates some common architectural components in recurrent sequence models and find it significantly improves the performance of both FoX and the Transformer. Our code is available at [`https://github.com/zhixuan-lin/forgetting-transformer`](https://github.com/zhixuan-lin/forgetting-transformer).

2025-01-21

ICLR.cc/2025/Conference (poster)

GenRL: Multimodal-foundation world models for generalization in embodied agents

Pietro Mazzaglia

Tim Verbelen

Bart Dhoedt

Sai Rajeswar

Learning generalist embodied agents, able to solve multitudes of tasks in different domains is a long-standing problem. Reinforcement learni… (see more)ng (RL) is hard to scale up as it requires a complex reward design for each task. In contrast, language can specify tasks in a more natural way. Current foundation vision-language models (VLMs) generally require fine-tuning or other adaptations to be adopted in embodied contexts, due to the significant domain gap. However, the lack of multimodal data in such domains represents an obstacle to developing foundation models for embodied applications. In this work, we overcome these problems by presenting multimodal-foundation world models, able to connect and align the representation of foundation VLMs with the latent space of generative world models for RL, without any language annotations. The resulting agent learning framework, GenRL, allows one to specify tasks through vision and/or language prompts, ground them in the embodied domain's dynamics, and learn the corresponding behaviors in imagination. As assessed through large-scale multi-task benchmarking in locomotion and manipulation domains, GenRL enables multi-task generalization from language and visual prompts. Furthermore, by introducing a data-free policy learning strategy, our approach lays the groundwork for foundational policy learning using generative world models. Website, code and data: https://mazpie.github.io/genrl/

2024-12-10

Conference on Neural Information Processing Systems (Accept (poster))

SAFT: Towards Out-of-Distribution Generalization in Fine-Tuning

Bac Nguyen

Stefan Uhlich

Fabien Cardinaux

Lukas Mauch

Marzieh Edraki

Handling distribution shifts from training data, known as out-of-distribution (OOD) generalization, poses a significant challenge in the fie… (see more)ld of machine learning. While a pre-trained vision-language model like CLIP has demonstrated remarkable zero-shot performance, further adaptation of the model to downstream tasks leads to undesirable degradation for OOD data. In this work, we introduce Sparse Adaptation for Fine-Tuning (SAFT), a method that prevents fine-tuning from forgetting the general knowledge in the pre-trained model. SAFT only updates a small subset of important parameters whose gradient magnitude is large, while keeping the other parameters frozen. SAFT is straightforward to implement and conceptually simple. Extensive experiments show that with only 0.1% of the model parameters, SAFT can significantly improve the performance of CLIP. It consistently outperforms baseline methods across several benchmarks. On the few-shot learning benchmark of ImageNet and its variants, SAFT gives a gain of 5.15% on average over the conventional fine-tuning method in OOD settings.

2024-12-06

Lecture Notes in Computer Science (published)

arxiv.org

SPARO: Selective Attention for Robust and Compositional Transformer Encodings for Vision

Ankit Vani

Bac Nguyen

Samuel Lavoie

Ranjay Krishna

Selective attention helps us focus on task-relevant aspects in the constant flood of our sensory input. This constraint in our perception al… (see more)lows us to robustly generalize under distractions and to new compositions of perceivable concepts. Transformers employ a similar notion of attention in their architecture, but representation learning models with transformer backbones like CLIP and DINO often fail to demonstrate robustness and compositionality. We highlight a missing architectural prior: unlike human perception, transformer encodings do not separately attend over individual concepts. In response, we propose SPARO, a read-out mechanism that partitions encodings into separately-attended slots, each produced by a single attention head. Using SPARO with CLIP imparts an inductive bias that the vision and text modalities are different views of a shared compositional world with the same corresponding concepts. Using SPARO, we demonstrate improvements on downstream recognition, robustness, retrieval, and compositionality benchmarks with CLIP (up to +14% for ImageNet, +4% for SugarCrepe), and on nearest neighbors and linear probe for ImageNet with DINO (+3% each). We also showcase a powerful ability to intervene and select individual SPARO concepts to further improve downstream task performance (up from +4% to +9% for SugarCrepe) and use this ability to study the robustness of SPARO's representation structure. Finally, we provide insights through ablation experiments and visualization of learned concepts.

2024-11-27

Lecture Notes in Computer Science (published)

arxiv.org

Not All LLM Reasoners Are Created Equal

Daniel Toyama

We study the depth of grade-school math (GSM) problem-solving capabilities of LLMs. To this end, we evaluate their performance on pairs of e… (see more)xisting math word problems together so that the answer to the second problem depends on correctly answering the first problem. Our findings reveal a significant reasoning gap in most LLMs, that is performance difference between solving the compositional pairs and solving each question independently. This gap is more pronounced in smaller, more cost-efficient, and math-specialized models. Moreover, instruction-tuning recipes and code generation have varying effects across LLM sizes, while finetuning on GSM can lead to task overfitting. Our analysis indicates that large reasoning gaps are not because of test-set leakage, but due to distraction from additional context and poor second-hop reasoning. Overall, LLMs exhibit systematic differences in their reasoning abilities, despite what their performance on standard benchmarks indicates.

2024-10-08

NeurIPS.cc/2024/Workshop/Sys2-Reasoning (poster)

VinePPO: Accurate Credit Assignment in RL for LLM Mathematical Reasoning

Amirhossein Kazemnejad

Nicolas 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.

2024-10-08

NeurIPS.cc/2024/Workshop/MATH-AI (accepted)

Adaptive Accompaniment with ReaLchords

Yusong Wu

Tim Cooijmans

Kyle Kastner

Adam Roberts

Ian Simon

Alexander Scarlatos

Chris Donahue

Cassie Tarakajian

Shayegan Omidshafiei

Natasha Jaques

Cheng-Zhi Anna Huang

Jamming requires coordination, anticipation, and collaborative creativity between musicians. Current generative models of music produce expr… (see more)essive output but are not able to generate in an \emph{online} manner, meaning simultaneously with other musicians (human or otherwise). We propose ReaLchords, an online generative model for improvising chord accompaniment to user melody. We start with an online model pretrained by maximum likelihood, and use reinforcement learning to finetune the model for online use. The finetuning objective leverages both a novel reward model that provides feedback on both harmonic and temporal coherency between melody and chord, and a divergence term that implements a novel type of distillation from a teacher model that can see the future melody. Through quantitative experiments and listening tests, we demonstrate that the resulting model adapts well to unfamiliar input and produce fitting accompaniment. ReaLchords opens the door to live jamming, as well as simultaneous co-creation in other modalities.

2024-07-22

International Conference on Machine Learning (Accept (Poster))