Back to Square One: Superhuman Performance in Chutes and Ladders Through Deep Neural Networks and Tree Search

2021/04/01 by Dylan R. Ashley, Ashley, Dylan, Anssi Kanervisto +4 · 1 voice
Computer Science · Economics, Econometrics and Finance · #Artificial Intelligence (cs.AI) #Artificial Intelligence in Games #Complex Systems and Time Series Analysis #FOS: Computer and information sciences #Sports Analytics and Performance #Time Series Analysis and Forecasting #cs.AI

paper · pdf · doi:10.48550/arxiv.2104.00698

openalex publication_date 2021/04/01 · arxiv published 2021/04/01 · arxiv updated 2021/04/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present AlphaChute: a state-of-the-art algorithm that achieves superhuman performance in the ancient game of Chutes and Ladders. We prove that our algorithm converges to the Nash equilibrium in constant time, and therefore is -- to the best of our knowledge -- the first such formal solution to this game. Surprisingly, despite all this, our implementation of AlphaChute remains relatively straightforward due to domain-specific adaptations. We provide the source code for AlphaChute here in our Appendix.

Citations

• Deep Mixtures of Factor Analysers
• Conditional Restricted Boltzmann Machines for Structured Output Prediction
• Discovering Multiple Constraints that are Frequently Approximately Satisfied
• Distilling the Knowledge in a Neural Network
• Efficient Parametric Projection Pursuit Density Estimation
• Deep Lambertian Networks
• Improving neural networks by preventing co-adaptation of feature detectors
• Training Products of Experts by Minimizing Contrastive Divergence
• A Fast Learning Algorithm for Deep Belief Nets
• Adaptive Mixtures of Local Experts
• Using Fast Weights to Attend to the Recent Past
• Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
• Distilling a Neural Network Into a Soft Decision Tree
• Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures
• Similarity of Neural Network Representations Revisited
• Cerberus: A Multi-headed Derenderer
• Learning Sparse Networks Using Targeted Dropout
• Modeling Documents with Deep Boltzmann Machines
• Large scale distributed neural network training through online distillation
• Lookahead Optimizer: k steps forward, 1 step back
• Subclass Distillation
• Teaching with Commentaries
• Unsupervised part representation by Flow Capsules

Discussions

• Superhuman Performance in Chutes and Ladders Through Deep Neural Networks [hn, 1 points, 0 comments]

Related

• Game Solving with Online Fine-Tuning
• Scenic trails ascending from sea-level Nim to alpine chess
• AlphaBeta is not as good as you think: a simple class of synthetic games for a better analysis of deterministic game-solving algorithms
• Discovering Expert-Level Nash Equilibrium Algorithms with Large Language Models
• Alphazzle: Jigsaw Puzzle Solver with Deep Monte-Carlo Tree Search