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

Although deep Reinforcement Learning (RL) has proven successful in a widerange of tasks, one challenge it faces is interpretability when applied toreal-world problems. Saliency maps are frequently used to provideinterpretability for deep neural networks. However, in the RL domain, existingsaliency map approaches are either computationally expensive and thus cannotsatisfy the real-time requirement of real-world scenarios or cannot produceinterpretable saliency maps for RL policies. In this work, we propose anapproach of Distillation with selective Input Gradient Regularization (DIGR)which uses policy distillation and input gradient regularization to produce newpolicies that achieve both high interpretability and computation efficiency ingenerating saliency maps. Our approach is also found to improve the robustnessof RL policies to multiple adversarial attacks. We conduct experiments on threetasks, MiniGrid (Fetch Object), Atari (Breakout) and CARLA Autonomous Driving,to demonstrate the importance and effectiveness of our approach.

Policy Distillation with Selective Input Gradient Regularization for Efficient Interpretability