English
Related papers

Related papers: On the Estimation Bias in Double Q-Learning

200 papers

Double machine learning is a statistical method for leveraging complex black-box models to construct approximately unbiased treatment effect estimates given observational data with high-dimensional covariates, under the assumption of a…

Machine Learning · Statistics 2022-06-03 Nitai Fingerhut , Matteo Sesia , Yaniv Romano

Double Q-learning is a classical control algorithm that mitigates the maximization bias of Q-learning. To do so, it explicitly trains two independent action-value functions and uses them to decouple action-selection and action-evaluation…

Machine Learning · Computer Science 2026-05-18 Prabhat Nagarajan , Martha White , Marlos C. Machado

Continuous control Deep Reinforcement Learning (RL) approaches are known to suffer from estimation biases, leading to suboptimal policies. This paper introduces innovative methods in RL, focusing on addressing and exploiting estimation…

Machine Learning · Computer Science 2024-10-14 Niccolò Turcato , Alberto Sinigaglia , Alberto Dalla Libera , Ruggero Carli , Gian Antonio Susto

In Reinforcement Learning the Q-learning algorithm provably converges to the optimal solution. However, as others have demonstrated, Q-learning can also overestimate the values and thereby spend too long exploring unhelpful states. Double…

Machine Learning · Computer Science 2023-03-16 David Barber

In classical Q-learning, the objective is to maximize the sum of discounted rewards through iteratively using the Bellman equation as an update, in an attempt to estimate the action value function of the optimal policy. Conventionally, the…

Machine Learning · Computer Science 2019-06-25 Hadi S. Jomaa , Josif Grabocka , Lars Schmidt-Thieme

We propose a new Q-learning variant, called 2RA Q-learning, that addresses some weaknesses of existing Q-learning methods in a principled manner. One such weakness is an underlying estimation bias which cannot be controlled and often…

Optimization and Control · Mathematics 2024-05-30 Peter Schmitt-Förster , Tobias Sutter

An agent's ability to leverage past experience is critical for efficiently solving new tasks. Prior work has focused on using value function estimates to obtain zero-shot approximations for solutions to a new task. In soft Q-learning, we…

Machine Learning · Computer Science 2024-06-27 Jacob Adamczyk , Volodymyr Makarenko , Stas Tiomkin , Rahul V. Kulkarni

Delusional bias is a fundamental source of error in approximate Q-learning. To date, the only techniques that explicitly address delusion require comprehensive search using tabular value estimates. In this paper, we develop efficient…

Machine Learning · Computer Science 2020-03-02 Andy Su , Jayden Ooi , Tyler Lu , Dale Schuurmans , Craig Boutilier

Fitted $Q$-iteration (FQI) and soft FQI are widely used value-based methods for offline reinforcement learning, but their standard stability guarantees often depend on Bellman completeness, a strong closure condition that can fail under…

Machine Learning · Statistics 2026-05-11 Lars van der Laan , Nathan Kallus

This paper discusses a new approach to the fundamental problem of learning optimal Q-functions. In this approach, optimal Q-functions are formulated as saddle points of a nonlinear Lagrangian function derived from the classic Bellman…

Machine Learning · Computer Science 2022-08-30 Huang Bojun

Overestimation in single-agent reinforcement learning has been extensively studied. In contrast, overestimation in the multiagent setting has received comparatively little attention although it increases with the number of agents and leads…

Multiagent Systems · Computer Science 2025-02-05 Yaodong Yang , Guangyong Chen , Hongyao Tang , Furui Liu , Danruo Deng , Pheng Ann Heng

We consider the problem of Approximate Dynamic Programming in relational domains. Inspired by the success of fitted Q-learning methods in propositional settings, we develop the first relational fitted Q-learning algorithms by representing…

Machine Learning · Computer Science 2020-06-11 Srijita Das , Sriraam Natarajan , Kaushik Roy , Ronald Parr , Kristian Kersting

Inspired by Double Q-learning algorithm, the Double-DQN (DDQN) algorithm was originally proposed in order to address the overestimation issue in the original DQN algorithm. The DDQN has successfully shown both theoretically and empirically…

Artificial Intelligence · Computer Science 2024-10-30 Shervin Halat , Mohammad Mehdi Ebadzadeh , Kiana Amani

Q-functions are widely used in discrete-time learning and control to model future costs arising from a given control policy, when the initial state and input are given. Although some of their properties are understood, Q-functions…

Optimization and Control · Mathematics 2019-02-21 Joseph Warrington

Reinforcement learning (RL) has seen significant research and application results but often requires large amounts of training data. This paper proposes two data-efficient off-policy RL methods that use parametrized Q-learning. In these…

Systems and Control · Electrical Eng. & Systems 2025-04-09 J. S. van Hulst , W. P. M. H. Heemels , D. J. Antunes

We study whether the learning rate $\alpha$, the discount factor $\gamma$ and the reward signal $r$ have an influence on the overestimation bias of the Q-Learning algorithm. Our preliminary results in environments which are stochastic and…

Machine Learning · Statistics 2022-10-12 Julius Wagenbach , Matthia Sabatelli

Overestimation bias control techniques are used by the majority of high-performing off-policy reinforcement learning algorithms. However, most of these techniques rely on pre-defined bias correction policies that are either not flexible…

Machine Learning · Computer Science 2022-02-01 Arsenii Kuznetsov , Alexander Grishin , Artem Tsypin , Arsenii Ashukha , Artur Kadurin , Dmitry Vetrov

Soft Q-learning is a variation of Q-learning designed to solve entropy regularized Markov decision problems where an agent aims to maximize the entropy regularized value function. Despite its empirical success, there have been limited…

Machine Learning · Computer Science 2024-09-06 Narim Jeong , Donghwan Lee

Guided exploration with expert demonstrations improves data efficiency for reinforcement learning, but current algorithms often overuse expert information. We propose a novel algorithm to speed up Q-learning with the help of a limited…

Machine Learning · Computer Science 2022-10-06 Fengdi Che , Xiru Zhu , Doina Precup , David Meger , Gregory Dudek

The breakthrough of deep Q-Learning on different types of environments revolutionized the algorithmic design of Reinforcement Learning to introduce more stable and robust algorithms, to that end many extensions to deep Q-Learning algorithm…

Machine Learning · Computer Science 2024-04-16 Mohammed Sabry , Amr M. A. Khalifa