arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2302.02563cond-mat.dis-nncs.LG
收藏

两层神经网络中由随机梯度下降诱导的表征漂移

Stochastic Gradient Descent-Induced Drift of Representation in a Two-Layer Neural Network

Cold Spring Harbor Laboratory(冷泉港实验室)

Farhad Pashakhanloo, Alexei Koulakov

AI总结:

本文针对两层线性前馈网络,在持续在线学习中分析SGD噪声诱导的表征漂移,通过法向/切空间分解得到涨落与扩散系数,并解释频繁刺激漂移更慢的实验现象。

摘要:

表征漂移是指在任务表现保持稳定的同时,神经激活随时间发生的变化。尽管在大脑和人工网络中都观察到了这一现象,但漂移的机制及其影响尚未得到充分理解。受最近在梨状皮层中发现的刺激依赖漂移的实验结果启发,我们利用理论和模拟研究两层线性前馈网络中的这一现象。具体而言,在持续在线学习场景中,我们研究随机梯度下降(SGD)内在噪声所诱导的漂移。通过将学习动力学分解到最小损失流形的法空间和切空间,我们证明前者对应于有限方差涨落,而后者可被视为流形上的有效扩散过程。我们解析地计算了隐藏层刺激表征的涨落和扩散系数,将其表示为网络参数和输入分布的函数。此外,与实验一致,我们表明刺激呈现得越频繁,漂移速率越慢。总体而言,我们的分析提供了一个理论框架,有助于更好地理解生物和人工神经网络中的漂移现象。

英文摘要:

Representational drift refers to over-time changes in neural activation accompanied by a stable task performance. Despite being observed in the brain and in artificial networks, the mechanisms of drift and its implications are not fully understood. Motivated by recent experimental findings of stimulus-dependent drift in the piriform cortex, we use theory and simulations to study this phenomenon in a two-layer linear feedforward network. Specifically, in a continual online learning scenario, we study the drift induced by the noise inherent in the Stochastic Gradient Descent (SGD). By decomposing the learning dynamics into the normal and tangent spaces of the minimum-loss manifold, we show the former corresponds to a finite variance fluctuation, while the latter could be considered as an effective diffusion process on the manifold. We analytically compute the fluctuation and the diffusion coefficients for the stimuli representations in the hidden layer as functions of network parameters and input distribution. Further, consistent with experiments, we show that the drift rate is slower for a more frequently presented stimulus. Overall, our analysis yields a theoretical framework for better understanding of the drift phenomenon in biological and artificial 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。

↑