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

Recent years have seen significant advances in explainable AI as the need tounderstand deep learning models has gained importance with the increasedemphasis on trust and ethics in AI. Comprehensible models for sequentialdecision tasks are a particular challenge as they require understanding notonly individual predictions but a series of predictions that interact withenvironmental dynamics. We present a framework for learning comprehensiblemodels of sequential decision tasks in which agent strategies are characterizedusing temporal logic formulas. Given a set of agent traces, we first clusterthe traces using a novel embedding method that captures frequent actionpatterns. We then search for logical formulas that explain the agent strategiesin the different clusters. We evaluate our framework on combat scenarios inStarCraft II (SC2), using traces from a handcrafted expert policy and a trainedreinforcement learning agent. We implemented a feature extractor for SC2environments that extracts traces as sequences of high-level featuresdescribing both the state of the environment and the agent's local behaviorfrom agent replays. We further designed a visualization tool depicting themovement of units in the environment that helps understand how different taskconditions lead to distinct agent behavior patterns in each trace cluster.Experimental results show that our framework is capable of separating agenttraces into distinct groups of behaviors for which our approach to strategyinference produces consistent, meaningful, and easily understood strategydescriptions.

A Framework for Understanding and Visualizing Strategies of RL Agents