An episodic intrinsic reward based on a contrastively learned temporal distance quasimetric improves exploration in sparse-reward Contextual MDPs.
TLDR : Unsupervised goal-conditioned RL via temporal distance-aware representations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Episodic Novelty Through Temporal Distance
An episodic intrinsic reward based on a contrastively learned temporal distance quasimetric improves exploration in sparse-reward Contextual MDPs.