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

Standard approaches to sequential decision-making exploit an agent's abilityto continually interact with its environment and improve its control policy.However, due to safety, ethical, and practicality constraints, this type oftrial-and-error experimentation is often infeasible in many real-world domainssuch as healthcare and robotics. Instead, control policies in these domains aretypically trained offline from previously logged data or in a growing-batchmanner. In this setting a fixed policy is deployed to the environment and usedto gather an entire batch of new data before being aggregated with past batchesand used to update the policy. This improvement cycle can then be repeatedmultiple times. While a limited number of such cycles is feasible in real-worlddomains, the quality and diversity of the resulting data are much lower than inthe standard continually-interacting approach. However, data collection inthese domains is often performed in conjunction with human experts, who areable to label or annotate the collected data. In this paper, we first explorethe trade-offs present in this growing-batch setting, and then investigate howinformation provided by a teacher (i.e., demonstrations, expert actions, andgradient information) can be leveraged at training time to mitigate the samplecomplexity and coverage requirements for actor-critic methods. We validate ourcontributions on tasks from the DeepMind Control Suite.

Knowledge Transfer from Teachers to Learners in Growing-Batch Reinforcement Learning