AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

Semantic HELM: A Human-Readable Memory for Reinforcement Learning

Fabian Paischer, Thomas Adler, Markus Hofmarcher, Sepp Hochreiter

NeurIPS 2023 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Language-grounded reinforcement agents · 5 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 semantic CLIP-token memory (SHELM) over memory-less PPOPPO · Avalon 10M stepsmean human normalized score: 0.146 -> 0.155No clear effect
SHELM vs HELMv2 on AvalonHELMv2 · Avalon 10M stepsmean human normalized score: 0.145 -> 0.155No clear effect
train memory-less PPO 50M steps (own) vs published PPOPPO · Avalon 50M stepsmean human normalized score: 0.165 -> 0.200Improved
memory-less PPO vs memory-based Dreamerv2 at 50MPPO · Avalon 50M stepsmean human normalized score: 0.200 vs 0.199No clear effect
replace semantic tokens with arbitrary retrieval after batch norm (HELMv2-Ret) (ablation)SHELM · MiniGrid-Memoryaccumulated rewardHurt
replace CNN current-observation branch with CLIP encoder (SHELM-CLIP) (ablation)SHELM · MiniGrid-Memoryaccumulated rewardHurt
use task-specific CNN features in history branch (SHELM-CNN) (ablation)SHELM · MiniGrid-Memoryaccumulated rewardHurt
replace pretrained TrXL LM with LSTM on text tokens (SHELM-LSTM) (ablation)SHELM · MiniGrid-Memoryaccumulated rewardHurt
swap ViT-B/16 CLIP encoder for RN50 (ablation)SHELM · MiniGrid-Memoryaccumulated rewardHurt
use designed prompts such as 'A screenshot of a tok' instead of 'An image of a tok' (ablation)SHELM token retrieval · Avalon and Psychlabtoken retrieval qualityImproved
use prompts on other environments (ablation)SHELM token retrieval · MiniGrid/MiniWorldtoken retrieval qualityNo clear effect
SHELM vs HELMv2 on Psychlab continuous recognitionHELMv2 · Psychlab CRrewardImproved

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