Travelled to:1 × China
1 × USA
Collaborated with:Bilal Kartal Pablo Hernandez-Leal P.Stone H.Bou-Ammar E.Eaton P.Ruvolo Chao Gao
Talks about:learn (6) reinforc (5) task (3) deep (3) auxiliari (2) pommerman (1) transfer (1) gradient (1) predict (1) guidanc (1)
Person: Matthew E. Taylor
DBLP: Taylor:Matthew_E=
Contributed to:
20142007
2019Wrote 6 papers:
- ICML-c2-2014-Bou-AmmarERT #learning #multi #online #policy
- Online Multi-Task Learning for Policy Gradient Methods (HBA, EE, PR, MET), pp. 1206–1214.
- ICML-2007-TaylorS #learning
- Cross-domain transfer for reinforcement learning (MET, PS), pp. 879–886.
- AIIDE-2019-GaoKHT #case study #learning #on the
- On Hard Exploration for Reinforcement Learning: A Case Study in Pommerman (CG, BK, PHL, MET), pp. 24–30.
- AIIDE-2019-Hernandez-LealK #learning #modelling
- Agent Modeling as Auxiliary Task for Deep Reinforcement Learning (PHL, BK, MET), pp. 31–37.
- AIIDE-2019-KartalHT #learning #predict
- Terminal Prediction as an Auxiliary Task for Deep Reinforcement Learning (BK, PHL, MET), pp. 38–44.
- AIIDE-2019-KartalHT19a #learning
- Action Guidance with MCTS for Deep Reinforcement Learning (BK, PHL, MET), pp. 153–159.