7 papers
Spatial Reasoning in LLM Game Agents: Impact of Causal Context and Multi-Step Planning
Mohit Jiwatode, Ronja Fuchs, Robin Schmöcker +2
LLM-based game agents often perform poorly on more complex tasks. This work examines whether these failures are linked to limited spatial reasoning and evaluates whether causal pro…
Discovering State Equivalences in UCT Search Trees By Action Pruning
Robin Schmöcker, Alexander Dockhorn, Bodo Rosenhahn
One approach to enhance Monte Carlo Tree Search (MCTS) is to improve its sample efficiency by grouping/abstracting states or state-action pairs and sharing statistics within a grou…
Grouping Nodes With Known Value Differences: A Lossless UCT-based Abstraction Algorithm
Robin Schmöcker, Alexander Dockhorn, Bodo Rosenhahn
A core challenge of Monte Carlo Tree Search (MCTS) is its sample efficiency, which can be improved by grouping state-action pairs and using their aggregate statistics instead of si…
Investigating Intra-Abstraction Policies For Non-exact Abstraction Algorithms
Robin Schmöcker, Alexander Dockhorn, Bodo Rosenhahn
One weakness of Monte Carlo Tree Search (MCTS) is its sample efficiency which can be addressed by building and using state and/or action abstractions in parallel to the tree search…
AUPO -- Abstracted Until Proven Otherwise: A Reward Distribution Based Abstraction Algorithm
Robin Schmöcker, Alexander Dockhorn, Bodo Rosenhahn
We introduce a novel, drop-in modification to Monte Carlo Tree Search's (MCTS) decision policy that we call AUPO. Comparisons based on a range of IPPC benchmark problems show that…
Investigating Scale Independent UCT Exploration Factor Strategies
Robin Schmöcker, Christoph Schnell, Alexander Dockhorn
The Upper Confidence Bounds For Trees (UCT) algorithm is not agnostic to the reward scale of the game it is applied to. For zero-sum games with the sparse rewards of a…