AI Paper Atlas

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

Discovery of Useful Questions as Auxiliary Tasks

Vivek Veeriah, Matteo Hessel, Zhongwen Xu, Janarthanan Rajendran, Richard L. Lewis, Junhyuk Oh, Hado P van Hasselt, David Silver, Satinder Singh

NeurIPS 2019 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Language-grounded reinforcement agents · 31 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
replace hand-crafted auxiliary tasks (reward prediction, pixel control, random GVFs) with meta-learned GVF questions, representation trained by auxiliary onlyIMPALA / A2C agent · Collect-Objects, Puddleworld, Atarimean episode returnImproved
use multi-step (non-myopic) meta-gradient with k>2 instead of myopic k=1 (ablation)Discovered GVFs · Collect-Objectsepisode returnImproved
increase unroll length from k=25 to k=50 (ablation)Discovered GVFs · Collect-Objectsfinal and average returnHurt
use too few questions (d=2) (ablation)Discovered GVFs · Collect-Objectsfinal returnHurt
use too many questions (d=128 vs smaller) (ablation)Discovered GVFs · Collect-Objectsaverage returnHurt
meta-loss as area under curve (summed) rather than end-batch only (ablation)Discovered GVFs · Atari (3 games)episode return stabilityImproved
add discovered GVF auxiliary tasks jointly to IMPALAIMPALA · Atari top-10 pixel-control gamesrelative performance gain: large gains in 6 of 10, small gains in 2, losses in 2Mixed
add reward prediction auxiliary task to IMPALAIMPALA · Atari 57median normalized scoreNo clear effect

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