AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Decision Stacks: Flexible Reinforcement Learning via Modular Generative Models

Siyan Zhao, Aditya Grover

NeurIPS 2023 · Reinforcement learning › Visual and representation RL · approach: In-context and adaptive RL · 0 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 reward modeling as input to the diffusion action model (ablation)Decision Stacks (diffusion obs) · Hopper-medium-v2 POMDPnormalized score: 43.7 -> 58.2Improved
Add reward modeling as input to the transformer action model (ablation)Decision Stacks (diffusion obs) · Hopper-medium-v2 POMDPnormalized score: 43.6 -> 57.7Improved
Add reward modeling as input to the MLP action model (ablation)Decision Stacks (diffusion obs) · Hopper-medium-v2 POMDPnormalized score: 38.1 -> 45.6Improved
Add reward modeling to transformer action model with transformer observation model (ablation)Decision Stacks (transformer obs) · Hopper-medium-v2 POMDPnormalized score: 45.1 -> 58.0Improved
Replace MLP action model with diffusion action model (ablation)Decision Stacks (diffusion obs) · Hopper-medium-v2 POMDPnormalized score: 45.6 -> 58.2Improved
Use hybrid diffusion obs / transformer reward / diffusion action instead of pure transformer (ablation)Decision Stacks · Hopper-medium-v2 POMDPnormalized score: 53.0 -> 58.2Improved
Use Decision Stacks modular action model instead of Decision Diffuser MLP inverse dynamicsDecision Diffuser · Maze2D Single Goal largenormalized score: 111.8 -> 171.6Improved
Use Decision Stacks instead of Diffuser without handcoded controllerDiffuser · Maze2D Multi Goal averagenormalized score: 70.0 -> 123.4Improved
Use Decision Stacks instead of Decision Diffuser on Maze2D single goal umazeDecision Diffuser · Maze2D Single Goal umazenormalized score: 113.8 -> 111.3No clear effect
Use Decision Stacks instead of Decision Diffuser in MDP locomotionDecision Diffuser · D4RL locomotion MDP averagenormalized score: 82.2 -> 81.1No clear effect
Use Decision Stacks instead of Diffuser in POMDP locomotionDiffuser · D4RL locomotion POMDP averagenormalized score: 64.2 -> 74.3Improved
Use Decision Stacks instead of Diffuser in MDP locomotionDiffuser · D4RL locomotion MDP averagenormalized score: 75.3 -> 81.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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