# Revisiting Overestimation Bias Problem of Q-learning: Settling Large Discrete Action Space via Action Intersection

arXiv: https://arxiv.org/abs/2608.12912 · HTML: https://mlsanity.com/paper/2608.12912/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL theory and sample complexity, one of 15 directions in Deep reinforcement learning methods.
- Builds on 14 of its 29 references that are in the map.
- First posted on arXiv 13 Aug 2026.

## Placed in

- Deep reinforcement learning methods / [RL theory and sample complexity](https://mlsanity.com/deeprl/rl-theory-and-sample-complexity/)

## Builds on (most cited first, 14 of 14)

- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Randomized Ensembled Double Q-Learning: Learning Fast Without a Model](https://mlsanity.com/paper/2101.05982/) (2021)
- [From r to Q^*: Your Language Model is Secretly a Q-Function](https://mlsanity.com/paper/2404.12358/) (2024)
- [Sample Efficient Actor-Critic with Experience Replay](https://mlsanity.com/paper/1611.01224/) (2016)
- [Deep Reinforcement Learning in Large Discrete Action Spaces](https://mlsanity.com/paper/1512.07679/) (2015)
- [Maxmin Q-learning: Controlling the Estimation Bias of Q-learning](https://mlsanity.com/paper/2002.06487/) (2020)
- [Averaged-DQN: Variance Reduction and Stabilization for Deep Reinforcement Learning](https://mlsanity.com/paper/1611.01929/) (2016)
- [Ensemble Bootstrapping for Q-Learning](https://mlsanity.com/paper/2103.00445/) (2021)
- [Sample Efficient Deep Reinforcement Learning for Dialogue Systems with Large Action Spaces](https://mlsanity.com/paper/1802.03753/) (2018)
- [Exact Reduction of Huge Action Spaces in General Reinforcement Learning](https://mlsanity.com/paper/2012.10200/) (2020)
- [Adaptive Ensemble Q-learning: Minimizing Estimation Bias via Error Feedback](https://mlsanity.com/paper/2306.11918/) (2023)
- [Action Candidate Based Clipped Double Q-learning for Discrete and Continuous Action Tasks](https://mlsanity.com/paper/2105.00704/) (2021)
- [On the Estimation Bias in Double Q-Learning](https://mlsanity.com/paper/2109.14419/) (2021)
- [M2DQN: A Robust Method for Accelerating Deep Q-learning Network](https://mlsanity.com/paper/2209.07809/) (2022)

## Abstract (arXiv)

This paper considers the overestimation bias problem of Q-learning in the setting of a large action space, for the purpose of relieving the bottleneck of existing methods. We find that the large action space increases the randomness in Q-value estimation. The randomness makes two paradigms that drive the major literature on the overestimation problem have their own bottlenecks: the coupling paradigm, i.e., the optimal action and its Q-value are estimated with the same Q-function, always has a positive bias. This is because randomness leads to some actions having abnormally high estimated values than their true values, and the coupling methods prefer these actions. The decoupling paradigm, i.e., the optimal action and its Q-value are estimated with two independent Q-functions, always has a negative bias. This is because randomness increases the estimation gap between the two independent Q-tables for the same action. This paper shows that action intersection can be a simple yet powerful strategy to relieve these bottlenecks. The action intersection strategy enables semi-decoupling via two designs: (1) it allows two Q-functions to share a certain fraction of trajectory data; (2) if a data sample is shared, each Q-function is updated using the coupling paradigm; otherwise, using the decoupling paradigm. Two properties make the action intersection strategy powerful: (1) attaining a large bias range, i.e., varying the data sharing fraction, the estimation bias varies from underestimating to overestimating; (2) fine granularity: the action intersection size can be made arbitrarily finer to enable finer control. We consider two experiment settings, i.e., tabular and deep RL, deep RL experiments show that our method outperforms several SOTA baselines drastically; tabular experiments reveal why our method can achieve superior performance.