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 2108.01210q-bio.NCcs.LG
收藏

基于Neural Data Transformers的神经群体活动表征学习

Representation learning for neural population activity with Neural Data Transformers

Georgia Institute of Technology(佐治亚理工学院);Emory University(埃默里大学)

Joel Ye, Chethan Pandarinath

AI总结:

针对循环神经网络建模神经群体动力学需顺序处理、速度慢的问题,提出非循环的Neural Data Transformer,建模效果媲美顶尖循环模型,推理速度快6倍以上,证明无需显式动力学模型也可建模神经群体自治动力学。

摘要:

理论认为,神经群体活动反映了一种潜在的动力学结构。使用具有显式动力学的状态空间模型(例如基于循环神经网络(RNN)的模型)可以准确捕获该结构。然而,使用循环机制显式建模动力学需要对数据进行顺序处理,会拖慢脑机接口等实时应用的速度。本文提出了一种非循环替代方案——Neural Data Transformer(NDT,神经数据Transformer)。我们通过将其应用于具有已知动力学的合成数据集,以及猴子在伸手任务中的运动皮层数据(该数据已被RNN很好地建模),测试了NDT捕获自治动力学系统的能力。NDT对这些数据集的建模效果与最先进的循环模型相当。此外,其非循环特性实现了3.9毫秒的推理速度,完全处于实时应用的循环时间范围内,在猴子伸手数据集上的速度比循环基线模型快6倍以上。这些结果表明,建模自治神经群体动力学并不一定需要显式动力学模型。代码地址:https://github.com/snel-repo/neural-data-transformers

英文摘要:

Neural population activity is theorized to reflect an underlying dynamical structure. This structure can be accurately captured using state space models with explicit dynamics, such as those based on recurrent neural networks (RNNs). However, using recurrence to explicitly model dynamics necessitates sequential processing of data, slowing real-time applications such as brain-computer interfaces. Here we introduce the Neural Data Transformer (NDT), a non-recurrent alternative. We test the NDT's ability to capture autonomous dynamical systems by applying it to synthetic datasets with known dynamics and data from monkey motor cortex during a reaching task well-modeled by RNNs. The NDT models these datasets as well as state-of-the-art recurrent models. Further, its non-recurrence enables 3.9ms inference, well within the loop time of real-time applications and more than 6 times faster than recurrent baselines on the monkey reaching dataset. These results suggest that an explicit dynamics model is not necessary to model autonomous neural population dynamics. Code: https://github.com/snel-repo/neural-data-transformers

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。

↑