arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2411.06403cs.AI
收藏

掌握NIM与无偏游戏的弱神经网络:一种受AlphaZero启发的多帧方法

Overcoming a Learning Bottleneck in Impartial Games: A Circuit Complexity Approach

Centre for Fundamentals of AI and Computational Theory, Queen Mary University of London(人工智能与计算理论基础研究中心,伦敦大学玛丽女王学院)

Søren Riis

AI总结:

本文提出一种受AlphaZero启发的多帧方法,通过多策略头和多帧架构解决NIM游戏中的不变量恢复问题,实现接近完美的位置分类。

摘要:

我们研究在固定延迟、固定规模量化推理(FSQI)条件下下的无偏游戏。在固定规模、有界范围的 regime 中,我们证明推理可以通过常深多项式大小的布尔电路(AC0)模拟。这导致了最坏情况的表示障碍:在FSQI/AC0 regime中,单帧代理无法强烈掌握NIM,因为最优玩法依赖于全局nim-sum(奇偶性)。在我们简化的确定性回放接口下,从结构化家族分析的单个回放策略头只能揭示不变量的一个固定线性函数,因此仅增加回放预算无法恢复缺失的位。我们推导出两个结构绕过:(1)一个多策略头回放架构,通过不同的回放通道恢复完整的不变量;(2)一个多帧架构,跟踪局部nimber差异并支持恢复。在多个设置中的实验与这些预测一致:单头基线保持在偶然水平,而两帧模型达到接近完美的恢复准确率,多头FSM控制的射击战达到完美胜利/失败位置分类。总体而言,实验结果支持了在FSQI/AC0 regime中显式结构先验(历史/差异或多个回放通道)的重要性。

英文摘要:

Why can neural networks struggle to learn optimal play in a game with a simple, efficient solution algorithm? Motivated by empirical difficulties with parity in impartial games, we use NIM to study a representational bottleneck and three ways around it. We show that a restricted model of shallow, fixed-scale quantised networks lies in $\mathrm{AC}^0$, a circuit class unable to compute parity on arbitrarily large inputs. This prevents optimal single-position play across all board sizes within the model. The obstruction also guides architectural design: two-frame inputs allow restoration from a known zero-nim-sum position; complementary rollout policies reveal the missing invariant; and persistent recurrence accumulates it row by row. Each route retains shallow network components while supplying additional information or processing time. Supervised experiments illustrate the benefits of two-frame restoration and multiple rollout heads; recurrence is a theoretical construction. These results establish representational possibilities, not guarantees of successful training: $\mathrm{AC}^0$ does not characterise practical learnability. Core mathematical results and constructions have been checked in Lean, with classical complexity results as explicit external inputs. The interactive edition and complete Lean sources are available online at https://sr123.github.io/impartial-games-learning-lean/.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑