AI Paper Atlas

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

Model-Free Trajectory-based Policy Optimization with Monotonic Improvement

Riad Akrour, Abbas Abdolmaleki, Hany Abdulsamad, Jan Peters, Gerhard Neumann

JMLR 2018 · Reinforcement learning › Policy gradient and actor-critic · approach: Model-based RL with guarantees · 19 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
use MOTO with entropy reduction constant beta0 = .1 instead of .5 (ablation)MOTO · Double link swing-upreturnImproved
replace GPS (linearized dynamics) with MOTO (quadratic Q-function)trajectory optimization · Double link swing-upreturnImproved
replace GPS with MOTO on a task with joint limitstrajectory optimization · Quad link swing-up (joint limits)returnImproved
add the entropy constraint (ablation)MOTO · Double link swing-up (M = 20)returnMixed
add importance sampling (ablation)MOTO · Double link swing-up (M = 20)convergence speedImproved
use dynamic programming Q-estimate instead of Monte-Carlo estimate (ablation)MOTO · Double link swing-up (M = 20)returnImproved
reduce rollouts per iteration M from 20 to 5 (ablation)MOTO · Double link swing-upsample efficiencyImproved
replace DMP-based REPS with MOTOrobot table tennis (no noise) · Robot table tennisconvergence speedImproved
replace DMP-based REPS/MORE with MOTOrobot table tennis · Robot table tennis (final policy quality)returnHurt
replace DMP-based methods with MOTO under bounce noiserobot table tennis · Robot table tennis (noisy bounce)returnImproved
replace default neural-network TRPO with MOTOmulti-link swing-up · Double and quad link swing-upreturnImproved
give TRPO a quadratic V-function (ablation)TRPO · Double link swing-upearly returnMixed

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