小柯机器人

美国哈佛大学Cengiz Pehlevan研究组取得一项新突破。他们发现噪声表征形成期间Hebbian/anti-Hebbian网络模型中感受野存在协调漂移。这一研究成果于2023年1月12日发表在国际学术期刊《自然-神经科学》上。

研究人员探讨神经表征优化了具有退化解空间的表征目标，并且嘈杂的突触更新诱导神经网络趋向于这个（接近）最佳空间，从而导致表征漂移的假设。研究说明了这个想法，并在简单的、生物学上合理的Hebbian/anti-Hebbian网络表征学习模型中探索其结果。研究发现单个神经元的漂移感受野可以通过协调的随机游走来表征，其有效扩散常数取决于各种参数，例如学习率、噪声幅度和输入统计量。尽管存在这种漂移，但总体代码的表征相似性随着时间推移相对稳定。该模型概括了对海马体和后顶叶皮层的实验观察结果，并做出了可行的预测，可以在未来实验中进行探测。

研究人员表示，最近的实验表明，即使动物已经完全学会并能稳定地执行任务，许多大脑区域的神经群体代码也会不断变化。这种代表性的“漂移”自然会产生对其原因、动态和功能的问题。

附：英文原文

Title: Coordinated drift of receptive fields in Hebbian/anti-Hebbian network models during noisy representation learning

Author: Qin, Shanshan, Farashahi, Shiva, Lipshutz, David, Sengupta, Anirvan M., Chklovskii, Dmitri B., Pehlevan, Cengiz

Issue&Volume: 2023-01-12

Abstract: Recent experiments have revealed that neural population codes in many brain areas continuously change even when animals have fully learned and stably perform their tasks. This representational ‘drift’ naturally leads to questions about its causes, dynamics and functions. Here we explore the hypothesis that neural representations optimize a representational objective with a degenerate solution space, and noisy synaptic updates drive the network to explore this (near-)optimal space causing representational drift. We illustrate this idea and explore its consequences in simple, biologically plausible Hebbian/anti-Hebbian network models of representation learning. We find that the drifting receptive fields of individual neurons can be characterized by a coordinated random walk, with effective diffusion constants depending on various parameters such as learning rate, noise amplitude and input statistics. Despite such drift, the representational similarity of population codes is stable over time. Our model recapitulates experimental observations in the hippocampus and posterior parietal cortex and makes testable predictions that can be probed in future experiments.

DOI: 10.1038/s41593-022-01225-z

Source: https://www.nature.com/articles/s41593-022-01225-z

Nature Neuroscience：《自然—神经科学》，创刊于1998年。隶属于施普林格·自然出版集团，最新IF：28.771
官方网址：https://www.nature.com/neuro/
投稿链接：https://mts-nn.nature.com/cgi-bin/main.plex