Goal-Driven Autonomy (GDA)
Goal Driven Autonomy (GDA) is a new goal reasoning method for problem solving in which autonomous agents dynamically identify and self-select their goals throughout execution. The objective of goal-driven autonomy (GDA) is to enable autonomous agents to direct the focus of their activities, and thus become more self-sufficient.
This project aims to study of Goal-driven autonomy (GDA) agents that are capable of learning expectations, explanations, and goals in such a way that they can (1) identify situations where discrepancies take place between what they expect and what actually happens, (2) explain the discrepancy, (3) decide which goals to achieve as a result of these explanations, and (4) act to accomplish these goals. These agents will be able to act in complex environments, which exhibit nondeterministic action outcomes and are dynamic with changes occurring independently of the agent’s own actions.
Personnel:
The following people have been involved with the project at various points:
Activities:
- We presented a paper at IJCAI-15
- We presented 3 papers at the ACS-15 Workshop on Advances in Cognitive Systems Workshop on Goal Reasoning
- We presented a paper at ICCBR-15
- Co-organized a workshop on Goal Reasoning: An alternate model of intelligent autonomy at the Advances in Cognitive Systems Conference (ACS-13). Here are the proceedings
- We have three papers accepted at ICCBR-13
- A news item about our NSF award
Publications:
- Dannenhauer, D. and Muñoz-Avila, H. (2015) Raising Expectations in GDA Agents Acting in Dynamic Environments. International Joint Conference on Artificial Intelligence (IJCAI-15). AAAI Press.
PDF
(a first answer to the question: is the GDA agent doing what its user programmed it to do?)
- Coman, A., Gillespie, H., Muñoz-Avila, H. and Cox, M. (2015) Believable Emotion-Influenced Perception: The Path to Motivated Rebel Agents. Advances in Cognitive Systems Workshop on Goal Reasoning)
PDF
- Dannenhauer, D., Muñoz-Avila, H. and Cox, M. (2015) Towards Cognition-level Goal Reasoning for Playing Real-Time Strategy Games. Advances in Cognitive Systems Workshop on Goal Reasoning)
PDF
-
Muñoz-Avila, H., Wilson, M. and Aha, D.W. (2015) Guiding the Ass with Goal Motivation Weights. Advances in Cognitive Systems Workshop on Goal Reasoning)
PDF
(learning motivators' weights. Motivators are used to choose the next goal to pursue)
- Dannenhauer, D. and Muñoz-Avila, H. (2015) Goal-Driven Autonomy with Semantically-annotated Hierarchical Cases. International Conference on Case-based Reasoning (ICCBR-15). Springer)
PDF
(integrating GDA, CBR and ontological reasoning)
- Coman, A. and Muñoz-Avila, H.
(2014) Motivation Discrepancies for Rebel Agents: Towards a Framework for Case-based Goal-Driven Autonomy for Character Believability. ICCBR-14 Workshop on Case-based Agents.
PDF
- Öztürk, P., Muñoz-Avila, H., Aamodt, A.
(2014) Explanation of Opportunities. ICCBR-14 Workshop on Case-based Agents.
PDF
- Dannenhauer, D. and Muñoz-Avila, H. (2013) LUIGi: A Goal-Driven Autonomy Agent Reasoning with Ontologies. Advances in Cognitive Systems Conference (ACS-13).
PDF
(integrating GDA and ontological reasoning)
- Ulit Jaidee, Héctor Muñoz-Avila, (2013) Modeling Unit Classes as Agents in Real-Time Strategy Games. The Ninth Annual AI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE-13). AAAI Press. Boston, MA
- Ulit Jaidee, Héctor Muñoz-Avila, David W. Aha. Case-Based Goal-Driven Coordination of Multiple Learning Agents. the twenty-first International Conference on Case-Based Reasoning (ICCBR). Saratoga, NY.
(addressing scalability issues in GDA systems by decomposing the learning and acting task among multiple agents)
- Giulio Finestrali, Héctor Muñoz-Avila. Case-Based Learning of Applicability Conditions for Stochastic Explanations. The twenty-first International Conference on Case-Based Reasoning (ICCBR). Saratoga, NY.
(introducing stochastic explanations in GDA)
- Dustin Dannenhauer, Hector Munoz-Avila. Case-based Goal Selection Inspired by IBM's Watson. the twenty-first International Conference on Case-Based Reasoning (ICCBR). Saratoga, NY.
(An architecture for goal selection inspired by IBM Watson)
- Jaidee, U., Muñoz-Avila, H., and Aha, D.W. (2012) Learning and Reusing Goal-Specific Policies for Goal-Driven Autonomy. International Conference on Case-based reasoning (ICCBR-12). Springer.
(enhances our work reported on IJCAI-11 by allowing lazy learning of policies within GDA)
- Jaidee, U., Muñoz-Avila, H., (2012) CLASSQL: A Q-Learning Algorithm for Adversarial Real-Time Strategy Games. AIIDE-12 Workshop on Artificial Intelligence in Adversarial Real-Time Games.
(focuses on a component of our GDA system and shows some level of coordination between agents)
- Jaidee, U., Muñoz-Avila, H., and Aha, D.W. (2011) Case-Based Learning in Goal-Driven Agents for Real-Time Strategy Combat Tasks. Proceedings ICCBR-2011 Workshop on Case-based reasoning for Computer Games.
(applying GDA for combat task scenarios in an RTS game)
- Jaidee, U., Munoz-Avila, H., Aha, D.W. (2011) Integrated Learning for Goal-Driven Autonomy. Proceedings of the Twenty-Second International Conference on Artificial Intelligence (IJCAI-11). AAAI Press.
(first GDA system to learn most GDA elements)
- Munoz-Avila, H., Aha, D.W., Jaidee, U., Carter, Elizabeth. (2010) Goal-Driven Autonomy with Case-Based Reasoning. Proceedings of the 18th International Conference on Case Based Reasoning (ICCBR 2010). Springer.
(Uses case-based reasoning techniques to reuse GDA knowledge)
- Muñoz-Avila, H. and. Aha, D.W. (2010) A Case Study of Goal-Driven Autonomy in Domination Games. AAAI-10 Workshop on Goal Directed Autonomy. AAAI Press.
(suggest ideas to use GDA in domination games. These ideas are developed in our ICCBR-10 paper)
- Munoz-Avila, H., Aha, D.W., Jaidee, U., Klenk, M., & Molineaux, M. (2010). Applying goal directed autonomy to a team shooter game. Proceedings of the Twenty-Third Florida Artificial Intelligence Research Society Conference. Daytona Beach, FL: AAAI Press.
(the first paper on GDA)
Contact author: Hector
Munoz-Avila
This material is based upon work supported by the National Science Foundation under Grant No. NSF 1217888. Any opinions, findings and conclusions or recomendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation (NSF)
Last updated: 7.20.2015