Unity Technologies has announced the open beta of Unity Machine Learning Agents (ML-Agents). Available later this year, Unity’s breakthrough AI toolkit will help enable machine learning developers and researchers to train agents in realistic, complex scenarios using Unity with decreased technical barriers than they could otherwise. This is critical future technology for many verticals, including robotics, automotive, and next-generation games. This first-of-its-kind advancement is in alignment with Unity’s mission to democratize access to superior technology and help developers solve hard problems.
“Machine learning is a disruptive technology that is important to all types of developers and researchers to make their games or systems smarter, but complexities and technical barriers make it out of reach for most,” said Danny Lange, Vice President of AI and Machine Learning at Unity Technologies. “This is an exciting new chapter in AI’s history as we are making an end-to-end machine learning environment widely accessible, and providing the critical tools needed to make more intelligent, beautiful games and applications. Complete with Unity’s physics engine and a 3D photorealistic rendering environment, our AI toolkit also offers a game-changing AI research platform to a rapidly growing community of AI enthusiasts exploring the frontiers of Deep Learning.”
ML- Agents, an open source toolkit, is specifically designed to help researchers and developers transform games and applications created using Unity into environments where intelligent agents can be trained. Using Reinforcement Learning, evolutionary strategies, and other machine learning methods through a simple to use Python API, ML-Agents has a superior advantage in solving complex machine learning problems in highly realistic environments.
The ML- Agents toolkit is adaptive and dynamic for a variety of use cases, including:
Academic researchers interested in studying complex multi-agent behavior in realistic competitive and cooperative scenarios.
Industry researchers interested in large-scale parallel training regimes for robotics, autonomous vehicle, and other industrial applications.
Game developers interested in filling virtual worlds with intelligent agents each acting with dynamic and engaging behavior.