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

Some of the most successful applications of deep reinforcement learning tochallenging domains in discrete and continuous control have used policygradient methods in the on-policy setting. However, policy gradients can sufferfrom large variance that may limit performance, and in practice requirecarefully tuned entropy regularization to prevent policy collapse. As analternative to policy gradient algorithms, we introduce V-MPO, an on-policyadaptation of Maximum a Posteriori Policy Optimization (MPO) that performspolicy iteration based on a learned state-value function. We show that V-MPOsurpasses previously reported scores for both the Atari-57 and DMLab-30benchmark suites in the multi-task setting, and does so reliably withoutimportance weighting, entropy regularization, or population-based tuning ofhyperparameters. On individual DMLab and Atari levels, the proposed algorithmcan achieve scores that are substantially higher than has previously beenreported. V-MPO is also applicable to problems with high-dimensional,continuous action spaces, which we demonstrate in the context of learning tocontrol simulated humanoids with 22 degrees of freedom from full stateobservations and 56 degrees of freedom from pixel observations, as well asexample OpenAI Gym tasks where V-MPO achieves substantially higher asymptoticscores than previously reported.

V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control