# Playing 20 Question Game with Policy-Based Reinforcement Learning

arXiv: https://arxiv.org/abs/1808.07645 · HTML: https://mlsanity.com/paper/1808.07645/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Aug 2018.

## Built on by (most cited first, 9 of 9)

- [Learning Neural Templates for Recommender Dialogue System](https://mlsanity.com/paper/2109.12302/) (2021)
- [Fantastic Rewards and How to Tame Them: A Case Study on Reward Learning for Task-oriented Dialogue Systems](https://mlsanity.com/paper/2302.10342/) (2023)
- [Active Sensing for Communications by Learning](https://mlsanity.com/paper/2112.04075/) (2021)
- [Learning to Ask: Conversational Product Search via Representation Learning](https://mlsanity.com/paper/2411.14466/) (2024)
- [KnowRL: Teaching Language Models to Know What They Know](https://mlsanity.com/paper/2510.11407/) (2025)
- [Probing the Lack of Stable Internal Beliefs in LLMs](https://mlsanity.com/paper/2603.25187/) (2026)
- [Converse, Focus and Guess -- Towards Multi-Document Driven Dialogue](https://mlsanity.com/paper/2102.02435/) (2021)
- [Sherlock Your Queries: Learning to Ask the Right Questions for Dialogue-Based Retrieval](https://mlsanity.com/paper/2510.18659/) (2025)
- [Query Your Way Out: Benchmarking Multi-Turn Knowledge Reasoning through Generalized Puzzle Solving](https://mlsanity.com/paper/2610.16133/) (2026)

## Abstract (arXiv)

The 20 Questions (Q20) game is a well known game which encourages deductive reasoning and creativity. In the game, the answerer first thinks of an object such as a famous person or a kind of animal. Then the questioner tries to guess the object by asking 20 questions. In a Q20 game system, the user is considered as the answerer while the system itself acts as the questioner which requires a good strategy of question selection to figure out the correct object and win the game. However, the optimal policy of question selection is hard to be derived due to the complexity and volatility of the game environment. In this paper, we propose a novel policy-based Reinforcement Learning (RL) method, which enables the questioner agent to learn the optimal policy of question selection through continuous interactions with users. To facilitate training, we also propose to use a reward network to estimate the more informative reward. Compared to previous methods, our RL method is robust to noisy answers and does not rely on the Knowledge Base of objects. Experimental results show that our RL method clearly outperforms an entropy-based engineering system and has competitive performance in a noisy-free simulation environment.