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

Approximation of the value functions in value-based deep reinforcementlearning induces overestimation bias, resulting in suboptimal policies. We showthat when the reinforcement signals received by the agents have a highvariance, deep actor-critic approaches that overcome the overestimation biaslead to a substantial underestimation bias. We first address the detrimentalissues in the existing approaches that aim to overcome such underestimationerror. Then, through extensive statistical analysis, we introduce a novel,parameter-free Deep Q-learning variant to reduce this underestimation bias indeterministic policy gradients. By sampling the weights of a linear combinationof two approximate critics from a highly shrunk estimation bias interval, ourQ-value update rule is not affected by the variance of the rewards received bythe agents throughout learning. We test the performance of the introducedimprovement on a set of MuJoCo and Box2D continuous control tasks anddemonstrate that it considerably outperforms the existing approaches andimproves the state-of-the-art by a significant margin.

Parameter-free Reduction of the Estimation Bias in Deep Reinforcement Learning for Deterministic Policy Gradients