Graphical models for interactive pomdps: Representations and solutions. In Eighth Internationl Conference on Autonomous Agents and Multiagents Systems Conference (AAMAS), pages 907-914, 2009. Improved approximation of interactive dynamic influence diagrams using discriminative model updates. In WIC/ACM/IEEE Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT), 2010.
![starcraft 2 mass recall starcraft 2 mass recall](https://1.bp.blogspot.com/-zfYJ97utD5k/Wtm5hWyyNHI/AAAAAAAAAMg/Ja2VQrNbPjAxG2uPxpsDNKSajOUtLbB1gCLcBGAs/s1600/Screenshot2018-04-20%2B00_47_37-min.jpg)
Epsilon-subject equivalence of models for interactive dynamic influence diagrams. In International Joint Conference on Artificial Intelligence (IJCAI), pages 39-45, 2015. Learning behaviors in agents systems with interactive dynamic influence diagrams. In Proceedings of the Fourteenth Internationl Conference on Autonomous Agents and Multiagents Systems Conference (AAMAS), pages 1161-1169, 2015. Iterative online planning in multiagent settings with limited model spaces and pac guarantees. Applied Artificial Intelligence: An International Journal, 23:855-871, 2009. Experiments with online reinforcement learning in real-time strategy games. In Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence (AAAI), pages 1995-2002, 2015. Scalable planning and learning for multiagent POMDPs: Extended version. We demonstrate the performance of the new technique in two problem domains. To achieve this, we propose to learn the value from available data particularly in practical applications of real-time strategy games. The new method reduces the space by additionally pruning behaviorally distinct models that result in the same expected value of the subject agent's optimal policy. In this paper, we challenge the minimal set of models and propose a value equivalence approach to further compress the model space. in order to minimize the model space, the previous I-DID techniques prune behaviorally equivalent models. The difficulty in solving I-DIDs is mainly due to an exponentially growing space of candidate models ascribed to other agents over time. They represent the problem of how a subject agent acts in a common setting shared with other agents who may act in sophisticated ways.
![starcraft 2 mass recall starcraft 2 mass recall](https://i.ytimg.com/vi/9iysH-H70Zc/hqdefault.jpg)
Interactive dynamic influence diagrams (I-DIDs) are recognized graphical models for sequential multiagent decision making under uncertainty.