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

Reinforcement Learning (RL) has become an increasingly important researcharea as the success of machine learning algorithms and methods grows. To combatthe safety concerns surrounding the freedom given to RL agents while training,there has been an increase in work concerning Safe Reinforcement Learning(SRL). However, these new and safe methods have been held to less scrutiny thantheir unsafe counterparts. For instance, comparisons among safe methods oftenlack fair evaluation across similar initial condition bounds and hyperparametersettings, use poor evaluation metrics, and cherry-pick the best training runsrather than averaging over multiple random seeds. In this work, we conduct anablation study using evaluation best practices to investigate the impact of runtime assurance (RTA), which monitors the system state and intervenes to assuresafety, on effective learning. By studying multiple RTA approaches in bothon-policy and off-policy RL algorithms, we seek to understand which RTA methodsare most effective, whether the agents become dependent on the RTA, and theimportance of reward shaping versus safe exploration in RL agent training. Ourconclusions shed light on the most promising directions of SRL, and ourevaluation methodology lays the groundwork for creating better comparisons infuture SRL work.

Ablation Study of How Run Time Assurance Impacts the Training and Performance of Reinforcement Learning Agents