Overview
This course teaches learners how to enhance any reinforcement learning algorithm by augmenting training data with techniques such as random crop, color jitter, patch cutout, and random convolutions. The course aims to improve sample efficiency, generalization to new environments, and overall performance of RL algorithms. The teaching method involves explaining the RAD module and its impact on RL algorithms, as presented in the provided paper. The intended audience for this course includes individuals interested in reinforcement learning, particularly those looking to improve the efficiency and performance of RL algorithms using data augmentation techniques.
Syllabus
Reinforcement Learning with Augmented Data (Paper Explained)
Taught by
Yannic Kilcher