Unsupervised Learning of Temporal Abstractions With Slot-Based Transformers
Sign Up to like & getrecommendations! Published in 2022 at "Neural Computation"
DOI: 10.48550/arxiv.2203.13573
Abstract: Abstract The discovery of reusable subroutines simplifies decision making and planning in complex reinforcement learning problems. Previous approaches propose to learn such temporal abstractions in an unsupervised fashion through observing state-action trajectories gathered from executing… read more here.
Keywords: learning temporal; based transformers; abstractions slot; temporal abstractions ... See more keywords