Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a detailed commentary on the research paper "Imagination-Augmented Agents for Deep Reinforcement Learning" in this 15-minute video. Dive into the novel architecture that combines model-free and model-based aspects of deep reinforcement learning. Discover how Imagination-Augmented Agents (I2As) learn to interpret predictions from a learned environment model to construct implicit plans. Understand the advantages of I2As, including improved data efficiency, performance, and robustness to model misspecification compared to existing baselines. Gain insights into the work of researchers from various institutions who contributed to this innovative approach in reinforcement learning.
Syllabus
Imagination-Augmented Agents for Deep Reinforcement Learning
Taught by
Yannic Kilcher