SURF: Semi-supervised reward learning with data augmentation for feedback-efficient preference-based reinforcement learning
Preference-based reinforcement learning (RL) has shown potential for teaching agents to
perform the target tasks without a costly, pre-defined reward function by learning the reward …
perform the target tasks without a costly, pre-defined reward function by learning the reward …
SURF: Semi-supervised Reward Learning with Data Augmentation for Feedback-efficient Preference-based Reinforcement Learning
J Park, Y Seo, J Shin, H Lee, P Abbeel… - arXiv e-prints, 2022 - ui.adsabs.harvard.edu
Preference-based reinforcement learning (RL) has shown potential for teaching agents to
perform the target tasks without a costly, pre-defined reward function by learning the reward …
perform the target tasks without a costly, pre-defined reward function by learning the reward …
SURF: Semi-supervised Reward Learning with Data Augmentation for Feedback-efficient Preference-based Reinforcement Learning
J Park, Y Seo, J Shin, H Lee, P Abbeel… - … Conference on Learning … - openreview.net
Preference-based reinforcement learning (RL) has shown potential for teaching agents to
perform the target tasks without a costly, pre-defined reward function by learning the reward …
perform the target tasks without a costly, pre-defined reward function by learning the reward …
SURF: SEMI-SUPERVISED REWARD LEARNING WITH DATA AUGMENTATION FOR FEEDBACK-EFFICIENT PREFERENCE-BASED REINFORCEMENT …
J Park, Y Seo, J Shin, H Lee, P Abbeel… - … Conference on Learning …, 2022 - pure.kaist.ac.kr
Preference-based reinforcement learning (RL) has shown potential for teaching agents to
perform the target tasks without a costly, pre-defined reward function by learning the reward …
perform the target tasks without a costly, pre-defined reward function by learning the reward …
SURF: Semi-supervised Reward Learning with Data Augmentation for Feedback-efficient Preference-based Reinforcement Learning
J Park, Y Seo, J Shin, H Lee, P Abbeel… - 10th International …, 2022 - koasas.kaist.ac.kr
Preference-based reinforcement learning (RL) has shown potential for teaching agents to
perform the target tasks without a costly, pre-defined reward function by learning the reward …
perform the target tasks without a costly, pre-defined reward function by learning the reward …
SURF: Semi-supervised Reward Learning with Data Augmentation for Feedback-efficient Preference-based Reinforcement Learning
J Park, Y Seo, J Shin, H Lee, P Abbeel… - Deep RL Workshop … - openreview.net
Preference-based reinforcement learning (RL) has shown potential for teaching agents to
perform the target tasks without a costly, pre-defined reward function by learning the reward …
perform the target tasks without a costly, pre-defined reward function by learning the reward …