Glad I got to join one even if part way through. Thanks for organizing!

A Method for Evaluating Hyperparameter Sensitivity in Reinforcement Learning The performance of modern reinforcement learning algorithms critically relies on tuning ever-increasing numbers of hyperparameters. Often, small changes in a hyperparameter can lead to drastic changes...

Paper now available: arxiv.org/abs/2412.07165

Paper soon to be arxiv. Will post when it’s up.

Our poster is Wednesday at 4:30pm, #6500. Talk to Jacob!

Is there a hyperparameters crisis in RL? And how do we measure and do something about it? With Jacob Adkins and Adam White. Talk to us at #NeurIPS2024.

Happy to be Local Co-Chair with Martha White! August is a great time to be in Edmonton. Plan to join us!

@ualberta.bsky.social