37 citations · 93 across the 12 of their papers we have counts for
12 papers
Graders should cheat: privileged information enables expert-level automated evaluations
Jin Peng Zhou, Sébastien M. R. Arnold, Nan Ding +3
Auto-evaluating language models (LMs), i.e., using a grader LM to evaluate the candidate LM, is an appealing way to accelerate the evaluation process and the cost associated with i…
Policy-Induced Self-Supervision Improves Representation Finetuning in Visual RL
Sébastien M. R. Arnold, Fei Sha
We study how to transfer representations pretrained on source tasks to target tasks in visual percept based RL. We analyze two popular approaches: freezing or finetuning the pretra…
A Domain-Agnostic Approach for Characterization of Lifelong Learning Systems
Megan M. Baker, Alexander New, Mario Aguilar-Simon +44
Despite the advancement of machine learning techniques in recent years, state-of-the-art systems lack robustness to "real world" events, where the input distributions and tasks enc…
Policy Learning and Evaluation with Randomized Quasi-Monte Carlo
Sebastien M. R. Arnold, Pierre L'Ecuyer, Liyu Chen +2
Reinforcement learning constantly deals with hard integrals, for example when computing expectations in policy evaluation and policy iteration. These integrals are rarely analytica…
Uniform Sampling over Episode Difficulty
Sébastien M. R. Arnold, Guneet S. Dhillon, Avinash Ravichandran +1
Episodic training is a core ingredient of few-shot learning to train models on tasks with limited labelled data. Despite its success, episodic training remains largely understudied…
Embedding Adaptation is Still Needed for Few-Shot Learning
Sébastien M. R. Arnold, Fei Sha
Constructing new and more challenging tasksets is a fruitful methodology to analyse and understand few-shot classification methods. Unfortunately, existing approaches to building t…
learn2learn: A Library for Meta-Learning Research
Sébastien M. R. Arnold, Praateek Mahajan, Debajyoti Datta +2
Meta-learning researchers face two fundamental issues in their empirical work: prototyping and reproducibility. Researchers are prone to make mistakes when prototyping new algorith…
Analyzing the Variance of Policy Gradient Estimators for the Linear-Quadratic Regulator
James A. Preiss, Sébastien M. R. Arnold, Chen-Yu Wei +1
We study the variance of the REINFORCE policy gradient estimator in environments with continuous state and action spaces, linear dynamics, quadratic cost, and Gaussian noise. These…
When MAML Can Adapt Fast and How to Assist When It Cannot
Sébastien M. R. Arnold, Shariq Iqbal, Fei Sha
Model-Agnostic Meta-Learning (MAML) and its variants have achieved success in meta-learning tasks on many datasets and settings. On the other hand, we have just started to understa…
Reducing the variance in online optimization by transporting past gradients
Sébastien M. R. Arnold, Pierre-Antoine Manzagol, Reza Babanezhad +2
Most stochastic optimization methods use gradients once before discarding them. While variance reduction methods have shown that reusing past gradients can be beneficial when there…
Shapechanger: Environments for Transfer Learning
Sébastien M. R. Arnold, Tsam Kiu Pun, Théo-Tim J. Denisart +1
We present Shapechanger, a library for transfer reinforcement learning specifically designed for robotic tasks. We consider three types of knowledge transfer---from simulation to s…
Accelerating SGD for Distributed Deep-Learning Using Approximated Hessian Matrix
Sébastien M. R. Arnold, Chunming Wang
We introduce a novel method to compute a rank approximation of the inverse of the Hessian matrix in the distributed regime. By leveraging the differences in gradients and param…