AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Training and Evaluation of Deep Policies Using Reinforcement Learning and Generative Models

Ali Ghadirzadeh, Petra Poklukar, Karol Arndt, Chelsea Finn, Ville Kyrki, Danica Kragic, Mårten Björkman

JMLR 2022 · Reinforcement learning › Policy gradient and actor-critic · approach: Regularized scalable deep value learning · 2 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
add a low beta (0.001) to the VAE generative model (ablation)GenRL · Hockey taskmax reward: -0.267 -> -0.040Improved
use VAE instead of InfoGAN as generative model (ablation)GenRL · Basketball taskmax reward: -0.081 vs -0.469Improved
use lambda 3.5 (stronger disentanglement) in InfoGAN (ablation)GenRL · Basketball taskmax reward: -1.183, -1.019, -1.344Hurt
higher recall of generative model (ablation)GenRL · Hockey task (VAE models)Pearson R with policy reward: 0.959Improved
higher disentanglement of generative model (ablation)GenRL · Hockey task (VAE models)Pearson R with policy reward: 0.979Improved
higher precision of generative model (ablation)GenRL · Hockey task (all models)Pearson R with policy reward: -0.068No clear effect
higher recall of generative model (ablation)GenRL · Basketball task (all models)Pearson R with policy reward: 0.907Improved
use GenRL instead of PPO and SACGenRL · Hockey and basketball tasksaverage reward during trainingImproved
use GenRL instead of PPO and SACGenRL · Hockey task safetytable-top collisions per episodeImproved

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