AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

What Can Learned Intrinsic Rewards Capture?

Zeyu Zheng, Junhyuk Oh, Matteo Hessel, Zhongwen Xu, Manuel Kroiss, Hado Van Hasselt, David Silver, Satinder Singh

ICML 2020 · Reinforcement learning › Visual and representation RL · approach: Reward shaping for sparse rewards · 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
use the learned intrinsic reward instead of extrinsic rewards or count-based explorationnew agents in Empty Rooms · Empty Roomsepisode returnImproved
use the learned intrinsic reward to explore uncertain object values and then exploit the better objectagents in Random ABC · Random ABCepisode returnImproved
use the learned intrinsic reward to discover the key-box relationshipagents in Key-Box · Key-Boxepisode returnImproved
use the learned intrinsic reward to prepare for recurring task changesagents in Non-stationary ABC · Non-stationary ABCepisode returnImproved
replace the lifetime-history reward input with current-timestep input (ablation)learned intrinsic reward · Empty Rooms, Random ABC, Key-Box, and Non-stationary ABCepisode returnHurt
replace the lifetime-return objective with the episodic-return objective (ablation)learned intrinsic reward · Random ABCepisode returnHurt
replace the lifetime-return objective with the episodic-return objective (ablation)learned intrinsic reward · Non-stationary ABCepisode returnHurt
replace the lifetime-return objective with the episodic-return objective (ablation)learned intrinsic reward · Key-Boxepisode returnHurt
replace the lifetime-return objective with the episodic-return objective (ablation)learned intrinsic reward · Empty Roomsepisode returnNo clear effect
train an agent using the learned intrinsic reward rather than a MAML-transferred policyagents in the paper's grid-world tasks · grid-world tasksepisode returnMixed
apply the learned intrinsic reward to agents with permuted or extended actionsagents in Random ABC · Random ABCepisode returnImproved
apply the actor-critic-trained intrinsic reward to a Q-learning agentQ-learning agent · Random ABCepisode returnImproved

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