发表机构
Korea University（高丽大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对逻辑门网络宽度扩展时核冗余导致性能饱和的问题，提出动态逻辑核框架，通过输入相关路由促进核组特化，提升核利用率和多样性，实现更高准确率与参数效率。
AI 中文摘要
可微逻辑门网络仅使用逻辑门进行操作，最近作为传统神经网络的高效替代方案引起了关注。然而，尽管其效率高，逻辑门网络的扩展行为仍未得到充分探索。相比之下，扩展模型容量是深度神经网络的核心设计原则，通常能带来性能提升。这一差异引出了一个关键问题：逻辑门网络能否也实现类似的扩展收益？在本工作中，我们聚焦于宽度作为主要扩展轴，并对其在逻辑门网络中的行为进行了系统分析。我们观察到，朴素的宽度扩展往往在逻辑核之间引入冗余，限制了额外核的有效利用，导致性能饱和。为解决这一局限，我们提出了一种动态逻辑核框架，通过促进核组间的特化来重新组织核利用。这使得网络能够通过输入相关的核路由更好地利用增加的宽度，同时确保路由和计算在推理时完全由门级布尔运算实现。我们进一步发现，核冗余在第一门级最为显著，这促使我们采用一种早期阶段的动态逻辑核策略，将适应集中在该层。实验结果表明，我们的方法提高了核利用率并增加了核多样性，从而在提升参数效率的同时实现了更高的准确率。
英文摘要
Differentiable logic gate networks, which operate using only logic gates, have recently attracted attention as an efficient alternative to conventional neural networks. However, despite their efficiency, the scaling behavior of logic gate networks remains underexplored. By contrast, scaling model capacity is a central design principle in deep neural networks and typically leads to improved performance. This discrepancy raises a key question: Can similar scaling benefits also be achieved in logic gate networks? In this work, we focus on width as a primary scaling axis and conduct a systematic analysis of its behavior in logic gate networks. We observe that naive width scaling often introduces redundancy among logic kernels, limiting the effective use of additional kernels and leading to performance saturation. To address this limitation, we propose a dynamic logic kernel framework that reorganizes kernel utilization by promoting specialization across kernel groups. This enables the network to better utilize increased width via input-dependent kernel routing, while ensuring that both routing and computation are implemented entirely with gate-level Boolean operations at inference time. We further find that kernel redundancy is most pronounced at the first gate level, motivating an early-stage dynamic logic kernel strategy that concentrates adaptation at this level. Experimental results demonstrate that our approach improves kernel utilization and increases kernel diversity, leading to higher accuracy with improved parameter efficiency.
CommentsNeurIPS 2026
