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

Text-based games have emerged as an important test-bed for ReinforcementLearning (RL) research, requiring RL agents to combine grounded languageunderstanding with sequential decision making. In this paper, we examine theproblem of infusing RL agents with commonsense knowledge. Such knowledge wouldallow agents to efficiently act in the world by pruning out implausibleactions, and to perform look-ahead planning to determine how current actionsmight affect future world states. We design a new text-based gaming environmentcalled TextWorld Commonsense (TWC) for training and evaluating RL agents with aspecific kind of commonsense knowledge about objects, their attributes, andaffordances. We also introduce several baseline RL agents which track thesequential context and dynamically retrieve the relevant commonsense knowledgefrom ConceptNet. We show that agents which incorporate commonsense knowledge inTWC perform better, while acting more efficiently. We conduct user-studies toestimate human performance on TWC and show that there is ample room for futureimprovement.

Text-based RL Agents with Commonsense Knowledge: New Challenges, Environments and Baselines