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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Preference-based reinforcement learning (RL) has shown potential for teachingagents to perform the target tasks without a costly, pre-defined rewardfunction by learning the reward with a supervisor's preference between the twoagent behaviors. However, preference-based learning often requires a largeamount of human feedback, making it difficult to apply this approach to variousapplications. This data-efficiency problem, on the other hand, has beentypically addressed by using unlabeled samples or data augmentation techniquesin the context of supervised learning. Motivated by the recent success of theseapproaches, we present SURF, a semi-supervised reward learning framework thatutilizes a large amount of unlabeled samples with data augmentation. In orderto leverage unlabeled samples for reward learning, we infer pseudo-labels ofthe unlabeled samples based on the confidence of the preference predictor. Tofurther improve the label-efficiency of reward learning, we introduce a newdata augmentation that temporally crops consecutive subsequences from theoriginal behaviors. Our experiments demonstrate that our approach significantlyimproves the feedback-efficiency of the state-of-the-art preference-basedmethod on a variety of locomotion and robotic manipulation tasks.

SURF: Semi-supervised Reward Learning with Data Augmentation for Feedback-efficient Preference-based Reinforcement Learning