DeepMind: Data-efficiency is one of the principal challenges for applying reinforcement learning on physical systems. We use hierarchical models to strengthen transfer while mitigating negative interference - saving weeks of training time for physical robots.
1 replies, 383 likes
Markus Wulfmeier: Proud to announce our recent work on compositional, hierarchical models to strengthen #transfer between related tasks while mitigating negative interference. We considerably improve #dataefficiency for reinforcement learning on physical #robots (reducing training time by weeks)
2 replies, 45 likes
Markus Wulfmeier 🏡: We'll be at the first virtual #RSS2020 later this year presenting RHPO ! Looking forward to many exciting conversations on transfer learning, multitask and obviously robots!
0 replies, 6 likes
Markus Wulfmeier 🏡: We have worked hard and here is the updated version of our work on compositional transfer learning in #ReinforcementLearning!
Thanks to all #RSS2020 (@RSS_Foundation ) reviewers for providing many helpful suggestions.
1 replies, 4 likes
Found on Jun 27 2019 at https://arxiv.org/pdf/1906.11228.pdf