Mastering diverse control tasks through world models by Danijar Hafner & Jurgis Pasukonis & Jimmy Ba & Timothy Lillicrap instant download

Mastering diverse control tasks through world models by Danijar Hafner & Jurgis Pasukonis & Jimmy Ba & Timothy Lillicrap instant download

Accepted: 5 February 2025Developing a general algorithm that learns to solve tasks across a wide range of Published online: 2 April 2025applications has been a fundamental challenge in artifcial intelligence. Although current reinforcement-learning algorithms can be readily applied to tasks similar to Open accesswhat they have been developed for, confguring them for new application domains Check for updatesrequires substantial human expertise and experimentation1,2. Here we present the third generation of Dreamer, a general algorithm that outperforms specialized methods across over 150 diverse tasks, with a single confguration. Dreamer learns a model of the environment and improves its behaviour by imagining future scenarios. Robustness techniques based on normalization, balancing and transformations enable stable learning across domains. Applied out of the box, Dreamer is, to our knowledge, the frst algorithm to collect diamonds in Minecraft from scratch without human data or curricula. This achievement has been posed as a substantial challenge in artifcial intelligence that requires exploring farsighted strategies from pixels and sparse rewards in an open world3. Our work allows solving challenging control problems without extensive experimentation, making reinforcement learning broadly applicable.