We will be covering topics of stochastic modelling and forecasting, Markov Decision Processes (MDP), and presenting new research on tackling large scale problems in stochastic and deterministic decision sciences. In this workshop we present flexible implementations of Monte Carlo Tree Search (MCTS), combined with domain specific knowledge and hybridization with other search algorithms. These can be powerful for finding the solutions to problems in complex planning. We introduce mctreesearch4j [1], an MCTS implementation written as a standard JVM library following key design principles of object oriented programming. We define key class abstractions allowing the MCTS library to flexibly adapt to any well defined Markov Decision Process or turn-based adversarial game. In addition, via the implementation of mctreesearch4j, the nuances of different types of MCTS algorithms are discussed.

Furthermore, we are in the process of inviting guest speakers to discuss topics on stochastic modelling, MDP’s, Multi-Armed Bandits, MCTS, reinforcement learning, and other related topics. The invited speakers will be presenting about state-of-the-art solvers to multi-armed bandit problems, as well as new advances in MCTS such as Neural MCTS.

During the end of the workshop, participants will apply their knowledge and demonstrate their newly built AI systems, and compete against each other in a benchmarked game environment. This cross-disciplinary Workshop combines the disciplines of management, computer science, economics, and artificial intelligence.

Contact Email: larkin.liu@tum.de
Please register at: https://www.eventbrite.ca/e/414129481427