Projects

Combinatorial AI is a learning algorithm implementing the principles of Machine Understanding elaborated in Egghead. The algorithm operates in the reinforcement learning paradigm and involves structural search and optimization.

Combinatorial AI is able to:

• Learn to build simple models of objects and processes in a changing environment
• Combine the simple models into solutions for problems defined in the environment
• Communicate the models using syntactically correct messages in a formal language
• Write syntactically valid programs in a formal language that solve the problems when executed

Combinatorial AI gains knowledge starting from very simple to more and more complex models.

Research continues on training Combinatorial AI to:

• Process structured problem descriptions in XML
• Represent solutions in a language that is Turing-complete

ELEPHANT is an open-source project on how to tain and use Machine Understanding. The method features:

• Step-by-step training from simple to complex
• Making use of automated hypotheses generation and trial
• Adapting to user-defined languages of subject domains
• The continued evolution of knowledge for solving problems

To make an impact, the principles of Machine Understanding need implementation in both training programs and learning agents. Once the training program for a subject domain is developed, it can be used, in principle, to train different agents.

Training programs that necessarily count on machine understanding can be used to benchmark different AI technologies.