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

Despite recent progress in reinforcement learning (RL), RL algorithms forexploration still remain an active area of research. Existing methods oftenfocus on state-based metrics, which do not consider the underlying causalstructures of the environment, and while recent research has begun to exploreRL environments for causal learning, these environments primarily leveragecausal information through causal inference or induction rather thanexploration. In contrast, human children - some of the most proficientexplorers - have been shown to use causal information to great benefit. In thiswork, we introduce a novel RL environment designed with a controllable causalstructure, which allows us to evaluate exploration strategies used by bothagents and children in a unified environment. In addition, throughexperimentation on both computation models and children, we demonstrate thatthere are significant differences between information-gain optimal RLexploration in causal environments and the exploration of children in the sameenvironments. We conclude with a discussion of how these findings may inspirenew directions of research into efficient exploration and disambiguation ofcausal structures for RL algorithms.

Learning Causal Overhypotheses through Exploration in Children and Computational Models