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 2106.05400q-bio.QMcs.LG
收藏

非线性神经网络的细粒度系统辨识

Fine-Grained System Identification of Nonlinear Neural Circuits

California Institute of Technology(加州理工学院)

Dawna Bagherian, James Gornet, Jeremy Bernstein, Yu-Li Ni, Yisong Yue, Markus Meister

摘要:

研究了高维组合函数的稀疏非线性模型恢复问题。我们的研究的动机是在神经科学中出现的机会,使用收集的测量数据恢复生物神经回路的细粒度模型。在神经科学领域知识的指导下,我们探索了可以恢复生成训练数据的潜在生物电路的条件。我们的研究结果提出了见解的理论和实际利益。最值得注意的是,我们发现,一个符号约束的权重是系统恢复的必要条件,我们建立了理论上的可识别性保证和经验上的模拟生物电路。最后,我们使用从小鼠视网膜收集的数据对视网膜神经节细胞回路进行了案例研究,展示了这种方法的实际潜力。

英文摘要:

We study the problem of sparse nonlinear model recovery of high dimensional compositional functions. Our study is motivated by emerging opportunities in neuroscience to recover fine-grained models of biological neural circuits using collected measurement data. Guided by available domain knowledge in neuroscience, we explore conditions under which one can recover the underlying biological circuit that generated the training data. Our results suggest insights of both theoretical and practical interests. Most notably, we find that a sign constraint on the weights is a necessary condition for system recovery, which we establish both theoretically with an identifiability guarantee and empirically on simulated biological circuits. We conclude with a case study on retinal ganglion cell circuits using data collected from mouse retina, showcasing the practical potential of this approach.

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。

↑