[2301.04104] Mastering Diverse Domains through World Models

General intelligence requires solving tasks across many domains. Current reinforcement learning algorithms carry this potential but are held back by the resources and knowledge required to tune them for new tasks. We present DreamerV3, a general and scalable algorithm based on world models that outperforms previous approaches across a wide range of domains with fixed hyperparameters. These domains include continuous and discrete actions, visual and low-dimensional inputs, 2D and 3D worlds, different data budgets, reward frequencies, and reward scales. We observe favorable scaling properties of DreamerV3, with larger models directly translating to higher data-efficiency and final performance. Applied out of the box, DreamerV3 is the first algorithm to collect diamonds in Minecraft from scratch without human data or curricula, a long-standing challenge in artificial intelligence. Our general algorithm makes reinforcement learning broadly applicable and allows scaling to hard decision making problems.

Mastering Diverse Domains through World Models Danijar Hafner,12 Jurgis Pasukonis,1 Jimmy Ba,2 Timothy Lillicrap1 Abstract Developing a general algorithm that learns to solve tasks across a wide range of arXiv:2301.04104v2 [cs.AI] 17 Apr 2024 applications has been a fundamental challenge in artificial intelligence. Although…

related reading

• [2509.24527] Training Agents Inside of Scalable World Modelsarxiv.org
• Introducing Dreamer: Scalable Reinforcement Learning Using World Modelsresearch.google
• A General Goal-Conditioned Minecraft Model - Pantographpantograph.com
• [2305.18290] Direct Preference Optimization: Your Language Model is Secretly a Reward Modelarxiv.org
• [2206.11795] Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videosarxiv.org
• [1709.06560] Deep Reinforcement Learning that Mattersarxiv.org
• DayDreamer: World Models for Physical Robot Learningdanijar.com
• [2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMsarxiv.org
• Dream to Control: Learning Behaviors by Latent Imaginationdanijar.com
• [2603.21972] Demystifying Reinforcement Learning for Long-Horizon Tool-Using Agents: A Comprehensive Recipearxiv.org
• [2606.05555] Representation Learning Enables Scalable Multitask Deep Reinforcement Learningarxiv.org
• Multi-task curriculum learning in a complex, visual,hard-exploration domain: Minecraftarxiv.org