Regret-Guided Search Control for Efficient Learning in AlphaZero

• URL: http://arxiv.org/abs/2602.20809v1
• Date: Tue, 24 Feb 2026 11:49:59 GMT
• Title: Regret-Guided Search Control for Efficient Learning in AlphaZero
• Authors: Yun-Jui Tsai, Wei-Yu Chen, Yan-Ru Ju, Yu-Hung Chang, Ti-Rong Wu,
• Abstract summary: Reinforcement learning (RL) agents achieve remarkable performance but remain far less learning-efficient than humans.<n>We propose Regret-Guided Search Control (RGSC), which extends AlphaZero with a regret network that learns to identify high-regret states.<n>RGSC provides an effective mechanism for search control, improving both efficiency and robustness of AlphaZero training.
• Score: 13.779557857453343
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning (RL) agents achieve remarkable performance but remain far less learning-efficient than humans. While RL agents require extensive self-play games to extract useful signals, humans often need only a few games, improving rapidly by repeatedly revisiting states where mistakes occurred. This idea, known as search control, aims to restart from valuable states rather than always from the initial state. In AlphaZero, prior work Go-Exploit applies this idea by sampling past states from self-play or search trees, but it treats all states equally, regardless of their learning potential. We propose Regret-Guided Search Control (RGSC), which extends AlphaZero with a regret network that learns to identify high-regret states, where the agent's evaluation diverges most from the actual outcome. These states are collected from both self-play trajectories and MCTS nodes, stored in a prioritized regret buffer, and reused as new starting positions. Across 9x9 Go, 10x10 Othello, and 11x11 Hex, RGSC outperforms AlphaZero and Go-Exploit by an average of 77 and 89 Elo, respectively. When training on a well-trained 9x9 Go model, RGSC further improves the win rate against KataGo from 69.3% to 78.2%, while both baselines show no improvement. These results demonstrate that RGSC provides an effective mechanism for search control, improving both efficiency and robustness of AlphaZero training. Our code is available at https://rlg.iis.sinica.edu.tw/papers/rgsc.

Related papers

• Impact of Decentralized Learning on Player Utilities in Stackelberg Games [57.08270857260131]
  In many two-agent systems, each agent learns separately and the rewards of the two agents are not perfectly aligned. We model these systems as Stackelberg games with decentralized learning and show that standard regret benchmarks result in worst-case linear regret for at least one player. We develop algorithms to achieve near-optimal $O(T2/3)$ regret for both players with respect to these benchmarks.
  arXiv  Detail & Related papers  (2024-02-29T23:38:28Z)
• REBEL: Reward Regularization-Based Approach for Robotic Reinforcement Learning from Human Feedback [61.54791065013767]
  A misalignment between the reward function and human preferences can lead to catastrophic outcomes in the real world.<n>Recent methods aim to mitigate misalignment by learning reward functions from human preferences.<n>We propose a novel concept of reward regularization within the robotic RLHF framework.
  arXiv  Detail & Related papers  (2023-12-22T04:56:37Z)
• A Coefficient Makes SVRG Effective [51.36251650664215]
  Variance Reduced Gradient (SVRG) is a theoretically compelling optimization method.<n>In this work, we demonstrate the potential of SVRG in optimizing real-world neural networks.
  arXiv  Detail & Related papers  (2023-11-09T18:47:44Z)
• AlphaZero Gomoku [9.434566356382529]
  We broaden the use of AlphaZero to Gomoku, an age-old tactical board game also referred to as "Five in a Row" Our tests demonstrate AlphaZero's versatility in adapting to games other than Go.
  arXiv  Detail & Related papers  (2023-09-04T00:20:06Z)
• Targeted Search Control in AlphaZero for Effective Policy Improvement [93.30151539224144]
  We introduce Go-Exploit, a novel search control strategy for AlphaZero. Go-Exploit samples the start state of its self-play trajectories from an archive of states of interest. Go-Exploit learns with a greater sample efficiency than standard AlphaZero.
  arXiv  Detail & Related papers  (2023-02-23T22:50:24Z)
• Are AlphaZero-like Agents Robust to Adversarial Perturbations? [73.13944217915089]
  AlphaZero (AZ) has demonstrated that neural-network-based Go AIs can surpass human performance by a large margin. We ask whether adversarial states exist for Go AIs that may lead them to play surprisingly wrong actions. We develop the first adversarial attack on Go AIs that can efficiently search for adversarial states by strategically reducing the search space.
  arXiv  Detail & Related papers  (2022-11-07T18:43:25Z)
• Does Zero-Shot Reinforcement Learning Exist? [11.741744003560095]
  A zero-shot RL agent is an agent that can solve any RL task instantly with no additional planning or learning. This marks a shift from the reward-centric RL paradigm towards "controllable" agents. Strategies for approximate zero-shot RL ave been suggested using successor features (SFs) or forward-backward (FB) representations.
  arXiv  Detail & Related papers  (2022-09-29T16:54:05Z)
• AlphaZero-Inspired General Board Game Learning and Playing [0.0]
  Recently, the seminal algorithms AlphaGo and AlphaZero have started a new era in game learning and deep reinforcement learning. In this paper, we pick an important element of AlphaZero - the Monte Carlo Tree Search (MCTS) planning stage - and combine it with reinforcement learning (RL) agents. We apply this new architecture to several complex games (Othello, ConnectFour, Rubik's Cube) and show the advantages achieved with this AlphaZero-inspired MCTS wrapper.
  arXiv  Detail & Related papers  (2022-04-28T07:04:14Z)
• Improve Agents without Retraining: Parallel Tree Search with Off-Policy Correction [63.595545216327245]
  We tackle two major challenges with Tree Search (TS) We first discover and analyze a counter-intuitive phenomenon: action selection through TS and a pre-trained value function often leads to lower performance compared to the original pre-trained agent. We introduce Batch-BFS: a GPU breadth-first search that advances all nodes in each depth of the tree simultaneously.
  arXiv  Detail & Related papers  (2021-07-04T19:32:24Z)
• Combining Deep Reinforcement Learning and Search for Imperfect-Information Games [30.520629802135574]
  We present ReBeL, a framework for self-play reinforcement learning and search provably converges to a Nash equilibrium in zero-sum games. We also show ReBeL achieves performance in heads-up no-limit Texas hold'em poker, while using far less domain knowledge than any prior poker AI.
  arXiv  Detail & Related papers  (2020-07-27T15:21:22Z)
• Learning to Prune Deep Neural Networks via Reinforcement Learning [64.85939668308966]
  PuRL is a deep reinforcement learning based algorithm for pruning neural networks. It achieves sparsity and accuracy comparable to current state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-07-09T13:06:07Z)
• Provable Self-Play Algorithms for Competitive Reinforcement Learning [48.12602400021397]
  We study self-play in competitive reinforcement learning under the setting of Markov games. We show that a self-play algorithm achieves regret $tildemathcalO(sqrtT)$ after playing $T$ steps of the game. We also introduce an explore-then-exploit style algorithm, which achieves a slightly worse regret $tildemathcalO(T2/3)$, but is guaranteed to run in time even in the worst case.
  arXiv  Detail & Related papers  (2020-02-10T18:44:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.