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

In this work we explore an auxiliary loss useful for reinforcement learningin environments where strong performing agents are required to be able tonavigate a spatial environment. The auxiliary loss proposed is to minimize theclassification error of a neural network classifier that predicts whether ornot a pair of states sampled from the agents current episode trajectory are inorder. The classifier takes as input a pair of states as well as the agent'smemory. The motivation for this auxiliary loss is that there is a strongcorrelation with which of a pair of states is more recent in the agents episodetrajectory and which of the two states is spatially closer to the agent. Ourhypothesis is that learning features to answer this question encourages theagent to learn and internalize in memory representations of states thatfacilitate spatial reasoning. We tested this auxiliary loss on a navigationtask in a gridworld and achieved 9.6% increase in accumulative episode rewardcompared to a strong baseline approach.

A Self-Supervised Auxiliary Loss for Deep RL in Partially Observable Settings