# IRumAI: Reinforcement Learning for Indian Rummy

arXiv: https://arxiv.org/abs/2606.21975 · HTML: https://mlsanity.com/paper/2606.21975/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods.
- Builds on 6 of its 14 references that are in the map.
- First posted on arXiv 20 Jun 2026.

## Placed in

- Deep reinforcement learning methods / [RL for games and combinatorial optimization](https://mlsanity.com/deeprl/rl-for-games-and-combinatorial-optimization/)

## Builds on (most cited first, 6 of 6)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [High-Dimensional Continuous Control Using Generalized Advantage Estimation](https://mlsanity.com/paper/1506.02438/) (2015)
- [DeepStack: Expert-Level Artificial Intelligence in No-Limit Poker](https://mlsanity.com/paper/1701.01724/) (2017)
- [PettingZoo: Gym for Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2009.14471/) (2020)
- [DouZero: Mastering DouDizhu with Self-Play Deep Reinforcement Learning](https://mlsanity.com/paper/2106.06135/) (2021)
- [Quantitative Rule-Based Strategy modeling in Classic Indian Rummy: A Metric Optimization Approach](https://mlsanity.com/paper/2601.00024/) (2025)

## Abstract (arXiv)

Despite its massive player base and complex hidden-information dynamics, Indian Rummy has received no reinforcement learning attention. Existing agents rely on combinatorial search, which is tactically strong but slow at inference. We present IRumAI, the first RL agent for the domain. IRumAI integrates Proximal Policy Optimization (PPO), meld-aware observation encoding, deadwood-driven reward shaping, and a dual-branch convolutional architecture. IRumAI is RL-trained solely against weak heuristics, after a one-time behaviour-cloning warm-start on stronger demonstration data. It generalises to defeat the entire baseline hierarchy, including a 53.9% win rate against the strongest search-based opponent unseen during RL training. Bypassing explicit search, IRumAI requires just 0.33 ms per action, which is over 7,000x faster than the state-of-the-art heuristic. Ablations validate our architectural choices, and linear probing reveals that the network implicitly models the opponent's hidden hand from public interactions.