AGMCTS augments MCTS with action-score gradients for particle beliefs, a Multiple Importance Sampling tree for reuse, and Area Formula gradients for smooth models, outperforming prior sample-based solvers on continuous benchmarks.
The cross-entropy method: a unified approach to combinatorial optimization, Monte-Carlo simulation and machine learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it