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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Exploration is a fundamental aspect of reinforcement learning (RL), and itseffectiveness is a deciding factor in the performance of RL algorithms,especially when facing sparse extrinsic rewards. Recent studies have shown theeffectiveness of encouraging exploration with intrinsic rewards estimated fromnovelties in observations. However, there is a gap between the novelty of anobservation and an exploration, as both the stochasticity in the environmentand the agent's behavior may affect the observation. To evaluate exploratorybehaviors accurately, we propose DEIR, a novel method in which we theoreticallyderive an intrinsic reward with a conditional mutual information term thatprincipally scales with the novelty contributed by agent explorations, and thenimplement the reward with a discriminative forward model. Extensive experimentson both standard and advanced exploration tasks in MiniGrid show that DEIRquickly learns a better policy than the baselines. Our evaluations on ProcGendemonstrate both the generalization capability and the general applicability ofour intrinsic reward. Our source code is available athttps://github.com/swan-utokyo/deir.

DEIR: Efficient and Robust Exploration through Discriminative-Model-Based Episodic Intrinsic Rewards