Abstract: I present our ongoing work on Goal-driven autonomy (GDA), a reflective model of goal reasoning that controls the focus of an agent’s planning activities by dynamically resolving unexpected discrepancies between the outcome of the agent’s actions and the agent’s own expectations. Discrepancies frequently arise when solving tasks in complex environments. GDA agents have performed well on such tasks by integrating methods for discrepancy recognition, explanation, goal formulation, and goal management. However, they require substantial domain knowledge, including, planning knowledge, knowledge about what constitutes a discrepancy and how to resolve it. We present work for acquiring this knowledge that integrates case-based reasoning, reinforcement learning and hierarchical task network learning methods. We assess its utility on tasks from a real-time strategy game. Taking advantage of hierarchical planning principles, we also present recent work eliciting expectations that not only validate the next action but the overall plan trajectory without requiring validation against the complete state.
Dr. Héctor Muñoz-Avila is currently serving as a program director at the National Science Foundation's (NSF) Robust Intelligence program. He is an associate professor at the Department of Computer Science and Engineering at Lehigh University.
Dr. Muñoz-Avila's is interested in intelligent systems that can assist people solving complex problems. This includes research on systems capable of acquiring and adapting episodic knowledge (i.e., case-based reasoning), generalizing and abstracting this episodic knowledge into general domain knowledge (i.e., machine learning), using this knowledge to solve new problems (i.e., planning), adapting to changes in the environment (i.e., agents) and introspectively reasoning about its own actions (i.e., cognitive systems). He is also interested in advancing game AI with AI techniques.
Dr. Muñoz-Avila is recipient of a National Science Foundation (NSF) CAREER award, two papers awards and held a Lehigh Class of 1961 Professorship. He has been chair for various international scientific meetings including the Sixth International Conference on Case-Based Reasoning (ICCBR-05) and the twenty-fifth Innovative Applications of AI Conference (IAAI-13). Dr. Muñoz-Avila is currently a program director for Robust Intelligence at the National Science Fundation (NSF). He has been funded in the past by the Defense Advanced Research Projects Agency (DARPA), the Office of Naval Research (ONR), and the Naval Research Laboratory (NRL) and the Air Force Research Laboratory (AFRL).
He holds a PhD (Dr. rer. nat.) in computer science from the Universität Kaiserslautern (Germany), an MS in computer science, a BS in Mathematics and a BS in computer Science from the Universidad de los Andes (Colombia).