法国巴黎文理研究大学N. Alex Cayco-Gajic等研究人员合作利用切片张量成分分析法降低神经子空间之外的维度。相关论文于2024年5月6日在线发表在《自然—神经科学》杂志上。

据了解，最近的研究表明，大尺度神经记录通常可以通过神经元间的共激活模式得到很好的描述。然而，认为神经变异性受限于固定的低维子空间的观点可能会忽略更高维的结构，包括刻板的神经序列或缓慢演化的潜空间。

研究人员认为神经数据中与任务相关的变异性也会随着试验或时间的推移而共同波动，从而定义出不同的“共变性类别”，这些类别可能会在同一数据集中同时出现。为了消除这些共变性类别，研究人员开发了sliceTCA（切片张量成分分析），这是一种针对神经数据张量的全新无监督降维方法。在三个示例数据集（包括灵长类动物经典伸手任务中的运动皮层活动和小鼠最近的多区域记录）中，结果表明，与传统方法相比，sliceTCA可以用较少的分量捕获神经数据中更多与任务相关的结构。

总之，这个理论框架扩展了低维群体活动的经典观点，并纳入了捕捉高维结构的额外潜变量类别。

附：英文原文

Title: Dimensionality reduction beyond neural subspaces with slice tensor component analysis

Author: Pellegrino, Arthur, Stein, Heike, Cayco-Gajic, N. Alex

Issue&Volume: 2024-05-06

Abstract: Recent work has argued that large-scale neural recordings are often well described by patterns of coactivation across neurons. Yet the view that neural variability is constrained to a fixed, low-dimensional subspace may overlook higher-dimensional structure, including stereotyped neural sequences or slowly evolving latent spaces. Here we argue that task-relevant variability in neural data can also cofluctuate over trials or time, defining distinct ‘covariability classes’ that may co-occur within the same dataset. To demix these covariability classes, we develop sliceTCA (slice tensor component analysis), a new unsupervised dimensionality reduction method for neural data tensors. In three example datasets, including motor cortical activity during a classic reaching task in primates and recent multiregion recordings in mice, we show that sliceTCA can capture more task-relevant structure in neural data using fewer components than traditional methods. Overall, our theoretical framework extends the classic view of low-dimensional population activity by incorporating additional classes of latent variables capturing higher-dimensional structure.

DOI: 10.1038/s41593-024-01626-2

Source: https://www.nature.com/articles/s41593-024-01626-2