conftrace_

2021 AAAI AAAI 2021

Extending Policy Shaping to Continuous State Spaces (Student Abstract)

Abstract

Abstract Policy Shaping is a Human-in-the-loop Reinforcement Learning (HRL) algorithm. We extend this work to continuous states with our algorithm, Deep Policy Shaping (DPS). DPS uses a feedback neural network that learns the optimality of actions from noisy feedback combined with an RL algorithm. In simulation, we find that DPS outperforms or matches baselines averaged over multiple hyperparameter settings and varying feedback correctness.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Reinforcement Learning

Authors

Thomas Wei, Taylor A. Kessler Faulkner, Andrea L. Thomaz

Topics

Reinforcement Learning > Methods > Deep RL Reinforcement Learning > Methods > Policy Learning Deep Learning > Learning Types > Reinforcement Learning Artificial Intelligence > Core AI > Reinforcement Learning

Keywords

deep reinforcement learning reinforcement learning human-in-the-loop learning deep learning continuous state space policy shaping human-in-the-loop reinforcement learning feedback neural network

More from AAAI 2021

All AAAI 2021 papers →

Chinese Character Image Clustering and Classification Based on Object Embedding Model (Student Abstract)

Is Each Layer Non-trivial in CNN? (Student Abstract)

Multi-label Few-shot Learning with Semantic Inference (Student Abstract)

Measuring Vegetation Density in Marsh Grass Photographs Using Deep Neural Networks (Student Abstract)

Remember More by Recalling Less: Investigating the Role of Batch Size in Continual Learning with Experience Replay (Student Abstract)

Towards Extracting Graph Neural Network Models via Prediction Queries (Student Abstract)