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 the numerous applications and success of deep reinforcement learningin many control tasks, it still suffers from many crucial problems andlimitations, including temporal credit assignment with sparse reward, absenceof effective exploration, and a brittle convergence that is extremely sensitiveto the hyperparameters of the problem. The problems of deep reinforcementlearning in continuous control, along with the success of evolutionaryalgorithms in facing some of these problems, have emerged the idea ofevolutionary reinforcement learning, which attracted many controversies.Despite successful results in a few studies in this field, a proper and fittingsolution to these problems and their limitations is yet to be presented. Thepresent study aims to study the efficiency of combining the two fields of deepreinforcement learning and evolutionary computations further and take a steptowards improving methods and the existing challenges. The "Evolutionary DeepReinforcement Learning Using Elite Buffer" algorithm introduced a novelmechanism through inspiration from interactive learning capability andhypothetical outcomes in the human brain. In this method, the utilization ofthe elite buffer (which is inspired by learning based on experiencegeneralization in the human mind), along with the existence of crossover andmutation operators, and interactive learning in successive generations, haveimproved efficiency, convergence, and proper advancement in the field ofcontinuous control. According to the results of experiments, the proposedmethod surpasses other well-known methods in environments with high complexityand dimension and is superior in resolving the mentioned problems andlimitations.

Evolutionary Deep Reinforcement Learning Using Elite Buffer: A Novel Approach Towards DRL Combined with EA in Continuous Control Tasks