Alexandre Piché — accepted AI-conference papers

2025

How to Train Your LLM Web Agent: A Statistical Diagnosis

Dheeraj Vattikonda, Santhoshi Ravichandran, Emiliano Penaloza, Hadi Nekoei, Thibault Le Sellier de Chezelles, Megh Thakkar +9

NeurIPS 2025poster

Large language model (LLM) agents for web interfaces have advanced rapidly, yet open-source systems still lag behind proprietary agents. Bridging this gap is key to enabling customizable, efficient, and privacy-preserving agents. Two challenges hinder progress: the reproducibility issues in RL and L…

Cited by 0 Source ↗Scholar ↗

2023

Mastering the Unsupervised Reinforcement Learning Benchmark from Pixels

Sai Rajeswar, Pietro Mazzaglia, Tim Verbelen, Alexandre Piché, Bart Dhoedt, Aaron Courville +1

ICML 2023oral

Controlling artificial agents from visual sensory data is an arduous task. Reinforcement learning (RL) algorithms can succeed but require large amounts of interactions between the agent and the environment. To alleviate the issue, unsupervised RL proposes to employ self-supervised interaction and le…

Cited by 25 Source ↗PDF ↗code ↗Scholar ↗

2021

Iterative Amortized Policy Optimization

Joseph Marino, Alexandre Piché, Alessandro Davide Ialongo, Yisong Yue

NeurIPS 2021poster

Policy networks are a central feature of deep reinforcement learning (RL) algorithms for continuous control, enabling the estimation and sampling of high-value actions. From the variational inference perspective on RL, policy networks, when used with entropy or KL regularization, are a form of amort…

Cited by 24 Source ↗PDF ↗code ↗Scholar ↗