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 2510.02765cs.LG
收藏

旋度下降:非梯度学习动态与符号多样的可塑性

Curl Descent: Non-Gradient Learning Dynamics with Sign-Diverse Plasticity

Département d’Études Cognitives(认知研究系);Edmond and Lily Safra Center for Brain Sciences(埃德蒙与莉莉·萨弗脑科学中心);École normale supérieure - PSL(巴黎高等师范学院-PSL);The Hebrew University of Jerusalem(耶路撒冷希伯来大学)

Hugo Ninou, Jonathan Kadmon, N. Alex Cayco-Gajic

AI总结:

本文探讨了生物神经网络学习过程中可能存在的非梯度'旋度'组件,通过分析抑制-兴奋连接网络和Hebbian/反Hebbian可塑性,发现非梯度动态在优化损失函数时的作用,揭示了特定架构支持鲁棒学习的机制。

摘要:

基于梯度的算法是人工神经网络训练的基石,但生物神经网络是否采用类似的梯度策略仍不明确。实验常发现突触可塑性规则的多样性,但这些是否是梯度下降的近似尚不清楚。本文研究了一种被忽视的可能性:学习动态可能包含根本上非梯度的'旋度'组件,同时仍能有效优化损失函数。旋度项自然出现在具有抑制-兴奋连接或Hebbian/反Hebbian可塑性的网络中,导致无法用任何目标的梯度下降来框架学习动态。为研究这些旋度项的影响,我们分析了可解析的学生-教师框架中的前馈网络,通过神经元表现出规则翻转的可塑性系统性引入非梯度动态。小旋度项保持原始解集的稳定性,导致学习动态类似于梯度下降。超过临界值后,强旋度项破坏解集稳定性。根据网络架构,这种稳定性丧失可能导致混沌学习动态,破坏性能。在其他情况下,旋度项可以反直觉地比梯度下降更快学习,通过允许权重动态暂时上升损失以逃离鞍点。本文结果识别了特定架构支持鲁棒学习的机制,为神经网络梯度基学习的规范理论提供了重要反驳。

英文摘要:

Gradient-based algorithms are a cornerstone of artificial neural network training, yet it remains unclear whether biological neural networks use similar gradient-based strategies during learning. Experiments often discover a diversity of synaptic plasticity rules, but whether these amount to an approximation to gradient descent is unclear. Here we investigate a previously overlooked possibility: that learning dynamics may include fundamentally non-gradient "curl"-like components while still being able to effectively optimize a loss function. Curl terms naturally emerge in networks with inhibitory-excitatory connectivity or Hebbian/anti-Hebbian plasticity, resulting in learning dynamics that cannot be framed as gradient descent on any objective. To investigate the impact of these curl terms, we analyze feedforward networks within an analytically tractable student-teacher framework, systematically introducing non-gradient dynamics through neurons exhibiting rule-flipped plasticity. Small curl terms preserve the stability of the original solution manifold, resulting in learning dynamics similar to gradient descent. Beyond a critical value, strong curl terms destabilize the solution manifold. Depending on the network architecture, this loss of stability can lead to chaotic learning dynamics that destroy performance. In other cases, the curl terms can counterintuitively speed learning compared to gradient descent by allowing the weight dynamics to escape saddles by temporarily ascending the loss. Our results identify specific architectures capable of supporting robust learning via diverse learning rules, providing an important counterpoint to normative theories of gradient-based learning in neural networks.

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。

↑