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

Optimal setting of several hyper-parameters in machine learning algorithms iskey to make the most of available data. To this aim, several methods such asevolutionary strategies, random search, Bayesian optimization and heuristicrules of thumb have been proposed. In reinforcement learning (RL), theinformation content of data gathered by the learning agent while interactingwith its environment is heavily dependent on the setting of manyhyper-parameters. Therefore, the user of an RL algorithm has to rely onsearch-based optimization methods, such as grid search or the Nelder-Meadsimplex algorithm, that are very inefficient for most RL tasks, slows downsignificantly the learning curve and leaves to the user the burden ofpurposefully biasing data gathering. In this work, in order to make an RLalgorithm more user-independent, a novel approach for autonomoushyper-parameter setting using Bayesian optimization is proposed. Data from pastepisodes and different hyper-parameter values are used at a meta-learning levelby performing behavioral cloning which helps improving the effectiveness inmaximizing a reinforcement learning variant of an acquisition function. Also,by tightly integrating Bayesian optimization in a reinforcement learning agentdesign, the number of state transitions needed to converge to the optimalpolicy for a given task is reduced. Computational experiments reveal promisingresults compared to other manual tweaking and optimization-based approacheswhich highlights the benefits of changing the algorithm hyper-parameters toincrease the information content of generated data.

Automatic tuning of hyper-parameters of reinforcement learning algorithms using Bayesian optimization with behavioral cloning