I

Ian Osband

Articles by Ian Osband

Academic · 1 min

Delightful Distributed Policy Gradient

arXiv:2603.20521v1 Announce Type: new Abstract: Distributed reinforcement learning trains on data from stale, buggy, or mismatched actors, producing actions with high surprisal (negative log-probability) under …

Ian Osband
14 views
Academic · 1 min

Does This Gradient Spark Joy?

arXiv:2603.20526v1 Announce Type: new Abstract: Policy gradient computes a backward pass for every sample, even though the backward pass is expensive and most samples carry …

Ian Osband
6 views