Deep Reinforcement Learning to Master the Asymmetric Strategy of Baghchal
Summary
This paper systematically explores four deep reinforcement learning solutions (DQN, REINFORCE, PPO, and MuZero) for the asymmetric Nepali board game Baghchal, finding that MuZero achieves the best win rates due to model-based planning via Monte Carlo Tree Search.
View Cached Full Text
Cached at: 07/22/26, 08:21 AM
# Deep Reinforcement Learning to Master the Asymmetric Strategy of Baghchal Source: [https://arxiv.org/abs/2607.18296](https://arxiv.org/abs/2607.18296) [View PDF](https://arxiv.org/pdf/2607.18296) > Abstract:Baghchal is a two\-player asymmetric board game with Nepali origins where four tigers are to capture goats and twenty goats desire to keep tigers in immobility\. Although Baghchal has a complex structure which is strategic, has perfect information structure, and has cultural meaning, it has not been adequately covered in deep reinforcement learning \(RL\) literature\. This paper gives a systematic exploration of four deep RL solutions Deep Q\-Network \(DQN\), REINFORCE, Proximal Policy Optimization \(PPO\) and MuZero that are trained on one side of the asymmetric gameplay of Baghchal and then evaluated on the other side\. The algorithms are rated based on win rate, draw rate, average captures, training convergence and computational cost\. It is experimentally found that MuZero generates the best performance in both tasks, achieving 86 percent win over these Tiger and 62 percent win over these Goat and the ability to do so is due to the model\-based planning machine through the Monte Carlo Tree Search\. PPO is the most realistic algorithm and is provided to be competitive over both asymmetric tasks with significantly reduced computational costs compared to MuZero\. Emergent strategic behavior analysis shows that model\-based strategies are optimal over long\-horizon planning, whereas value\-based counterparts like DQN are more biased up towards the Tiger role owing to the more substantial reward signal\. ## Submission history From: Ranjit Raut \[[view email](https://arxiv.org/show-email/55be6aee/2607.18296)\] **\[v1\]**Thu, 2 Jul 2026 14:22:42 UTC \(351 KB\)
Similar Articles
Learning Montezuma’s Revenge from a single demonstration
OpenAI demonstrates a method for training a reinforcement learning agent to play Montezuma's Revenge from a single human demonstration, addressing the challenge of sparse rewards through curriculum learning and careful hyperparameter tuning. The approach achieves strong performance on the notoriously difficult Atari game while showing generalization limitations on other titles.
From Imitation to Interaction: Mastering Game of Schnapsen with Shallow Reinforcement Learning
This paper investigates whether shallow neural network agents can master the card game Schnapsen using reinforcement learning, outperforming a supervised imitation baseline and achieving competitive results against a strong search-based opponent.
Discovering Lattice Reduction Strategies via Self-Play
This paper presents Delta-Star, a deep reinforcement learning approach using AlphaZero-style self-play to discover superior lattice reduction strategies by interacting with the primitive actions of the LLL algorithm. The learned policy generalizes to higher dimensions and unseen moduli without retraining.
Self-Play Reinforcement Learning under Imperfect Information in Big 2
This paper presents a self-play reinforcement learning framework for the four-player imperfect-information card game Big 2, comparing policy-gradient and value-based methods and finding that PPO with entropy regularization outperforms others.
Representation Learning Enables Scalable Multitask Deep Reinforcement Learning
This paper argues that representation learning, not model-based planning, is the key to scalable multitask deep reinforcement learning. It introduces MR.Q, a simple model-free algorithm with auxiliary predictive objectives that outperforms prior world-model-based methods across diverse continuous control tasks.