PAGI World (“pay-guy”), which stands for Psychometric Artificial General Intelligence World, is a simulation environment written in Unity 2D which allows AI and AGI researchers to test out their ideas on a wide variety of tasks, and to create their own tasks, in a real-time environment with realistic physics. PAGI World is designed to be a platform which many researchers can use to compare competing approaches to AI and AGI. With that goal in mind, there are many features which make PAGI World unique:
- Written in Unity 2D, so that updates to Unity 2D’s physics engine will be incorporated into PAGI World as they come out
- Many real-world elements have been created that don’t exist in other simulation environments, such as: Water/Fluid Dynamics, Sticky objects, Friction, Vision, etc.
- Can be run on any major operating system—Windows, Mac OS, most Linux distros, iOS, Android, and more
- Uses TCP/IP communication, so that AI agents can be written in virtually any programming language
- Has an easy-to-use GUI Task Builder, so that researchers can create their own PAI or PAGI tasks—or just their own environments for their AI agents to play in—without writing a single line of code.
- Licato, J. & Bringsjord, S. (2016). A Physically Realistic, General-Purpose Simulation Environment for Developmental AI Systems. In Proceedings of the ECAI 2016 Workshop on Evaluating General-Purpose AI (EGPAI 2016). - Use this as the main citation for PAGI World
- Atkin, K., Licato, J., & Bringsjord, S. (2015). Modeling Interoperability Between a Reflex and Reasoning System in a Physical Simulation Environment. In Proceedings of the 2015 Spring Simulation Multi-Conference.
- Bello, P., Licato, J., & Bringsjord, S. (2015). Constraints on Freely Chosen Action for Moral Robots: Consciousness and Control. In Proceedings of RO-MAN 2015.
- Bringsjord, S., Licato, J., Govindarajulu, N. S., Ghosh, R., & Sen, A. (2015). Real Robots that Pass Human Tests of Self-Consciousness. In Proceedings of RO-MAN 2015.
- Licato, J. & Bringsjord, S. (2015). PAGI World: A Simulation Environment to Challenge Cognitive Architectures. In IBM Cognitive Systems Institute Group Speaker Series.
- Marton, N., Licato, J., & Bringsjord, S. (2015). Creating and Reasoning Over Scene Descriptions in a Physically Realistic Simulation. In Proceedings of the 2015 Spring Simulation Multi-Conference.
- Licato, J., Marton, N., Dong, B., Sun, R., & Bringsjord, S. (2015). Modeling the Creation and Development of Cause-Effect Pairs for Explanation Generation in a Cognitive Architecture. In Proceedings of the 2015 International Workshop on Artificial Intelligence and Cognition (AIC 2015).