Data-Driven Goal Generation for Integrated Cognitive Systems

We describe our Meta-cognitive, Integrated, Dual-Cycle Architecture (MIDCA), whose purpose is to provide agents with a greater capacity for acting in an open world and dealing with unexpected events. We present MIDCA 1.0, a partial implementation which explores a novel machine-learning approach to goal generation using the Tilde and FOIL algorithms. We describe results from this goal generation algorithm and pre- view MIDCA 1.1, a partially implemented version that will guide the goal insertion process using statistical anomaly detection and categorization techniques. Finally, we outline the next steps towards a complete, functional implementation of the MIDCA architecture

Read more here: https://cdn.aaai.org/ocs/7618/7618-32589-1-PB.pdf