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

Reinforcement learning (RL) has achieved remarkable performance in numeroussequential decision making and control tasks. However, a common problem is thatlearned nearly optimal policy always overfits to the training environment andmay not be extended to situations never encountered during training. Forpractical applications, the randomness of environment usually leads to somedevastating events, which should be the focus of safety-critical systems suchas autonomous driving. In this paper, we introduce the minimax formulation anddistributional framework to improve the generalization ability of RL algorithmsand develop the Minimax Distributional Soft Actor-Critic (Minimax DSAC)algorithm. Minimax formulation aims to seek optimal policy considering the mostsevere variations from environment, in which the protagonist policy maximizesaction-value function while the adversary policy tries to minimize it.Distributional framework aims to learn a state-action return distribution, fromwhich we can model the risk of different returns explicitly, therebyformulating a risk-averse protagonist policy and a risk-seeking adversarialpolicy. We implement our method on the decision-making tasks of autonomousvehicles at intersections and test the trained policy in distinct environments.Results demonstrate that our method can greatly improve the generalizationability of the protagonist agent to different environmental variations.

Improving Generalization of Reinforcement Learning with Minimax Distributional Soft Actor-Critic