AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Exploring Model-based Planning with Policy Networks

Tingwu Wang, Jimmy Ba

ICLR 2020 · Reinforcement learning › Visual and representation RL · approach: RL for combinatorial routing · 80 citations (OpenAlex)

Read the paper

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
plan by perturbing policy parameters instead of action sequencesPOPLIN · Antreturn: 1165.5 ± 226.9 -> 2330.1 ± 320.9Improved
plan by perturbing policy parameters instead of action sequencesPOPLIN · Cheetahreturn: 4204.5 ± 789.0 -> 12227.9 ± 5652.8Improved
use POPLIN-A action-space planning rather than PETSmodel-based planning · Cheetahreturn: 4204.5 ± 789.0 -> 4651.1 ± 1088.5No clear effect
use POPLIN-P parameter-space planning rather than POPLIN-APOPLIN · Hopperreturn: 202.5 ± 962.5 -> 2055.2 ± 613.8Improved
use POPLIN-P parameter-space planning rather than POPLIN-APOPLIN · Pusherreturn: -56.0 ± 24.3 -> -55.8 ± 23.1No clear effect
use POPLIN-P parameter-space planning rather than POPLIN-APOPLIN · Pendulumreturn: 178.3 ± 19.3 -> 167.9 ± 45.9Hurt
replan the policy throughout each candidate trajectory instead of using the initialized reference sequence (ablation)POPLIN-A · MuJoCo benchmark environmentsreturnHurt
train POPLIN-A with imaginary planned data instead of only real data (ablation)POPLIN-A · MuJoCo benchmark environmentsreturnNo clear effect
use separate parameter noise at each time-step instead of shared noise (ablation)POPLIN-P · MuJoCo benchmark environmentsreturnImproved
distill the planned behavior with behavior cloning rather than parameter averaging (ablation)POPLIN-P policy control · CheetahreturnImproved
distill planned behavior with GAN rather than behavior cloning (ablation)POPLIN-P policy control · Cheetah and Cheetah-v0returnMixed
increase initial parameter-noise variance (ablation)POPLIN-P-Avg · MuJoCo benchmark environmentsreturnImproved

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