English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Complex sequential tasks in continuous-control settings often require agentsto successfully traverse a set of "narrow passages" in their state space.Solving such tasks with a sparse reward in a sample-efficient manner poses achallenge to modern reinforcement learning (RL) due to the associatedlong-horizon nature of the problem and the lack of sufficient positive signalduring learning. Various tools have been applied to address this challenge.When available, large sets of demonstrations can guide agent exploration.Hindsight relabelling on the other hand does not require additional sources ofinformation. However, existing strategies explore based on task-agnostic goaldistributions, which can render the solution of long-horizon tasks impractical.In this work, we extend hindsight relabelling mechanisms to guide explorationalong task-specific distributions implied by a small set of successfuldemonstrations. We evaluate the approach on four complex, single and dual arm,robotics manipulation tasks against strong suitable baselines. The methodrequires far fewer demonstrations to solve all tasks and achieves asignificantly higher overall performance as task complexity increases. Finally,we investigate the robustness of the proposed solution with respect to thequality of input representations and the number of demonstrations.

Wish you were here: Hindsight Goal Selection for long-horizon dexterous manipulation