AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

Hybrid Reinforcement Learning with Expert State Sequences

Xiaoxiao Guo, Shiyu Chang, Mo Yu, Gerald Tesauro, Murray Campbell

AAAI 2019 · Reinforcement learning › Imitation learning · approach: Imitation combined with RL · 23 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Combine A2C with imitation from inferred expert actionsHybrid RL · Taxicumulative rewardImproved
Use the dual tensor action inference model instead of an MLPHybrid RL and behavioral cloning · Taxicumulative rewardImproved
Add RL reward signals to behavioral cloningHybrid RL · Taxicumulative rewardImproved
Use rank 1 instead of rank 2 for the tensor action inference model (ablation)Hybrid RL · Taxicumulative rewardHurt
Increase tensor rank from 2 to 4 (ablation)Hybrid RL · Taxicumulative rewardNo clear effect
Use the hybrid objective instead of pure A2CHybrid RL · Aliengame scoreImproved
Use the hybrid objective instead of pure A2CHybrid RL · BeamRidergame scoreImproved
Use the hybrid objective instead of pure A2CHybrid RL · Breakoutgame scoreImproved
Use the hybrid objective instead of pure A2CHybrid RL · MsPacmangame scoreImproved
Use the hybrid objective instead of pure A2CHybrid RL · Ponggame scoreImproved
Use the hybrid objective instead of pure A2CHybrid RL · Qbertgame scoreImproved
Use the hybrid objective instead of pure A2CHybrid RL · Seaquestgame scoreNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy