[论文解读] How brains are built: Principles of computational neuroscience

Richard Granger|arXiv (Cornell University)|Mar 29, 2017

Neural dynamics and brain function参考文献 96被引用 5

一句话总结

本文提出，计算神经科学通过从第一性原理出发模拟神经系统，实现对脑功能的深层理解，将计算视为对底层机制的研究，而非硬件本身。它认为，完全模拟脑过程——类似于构建一个功能性的替代品——是理解的终极检验，对科学洞察力和类脑工程均具有重要意义。

ABSTRACT

'If I cannot build it, I do not understand it.' So said Nobel laureate Richard Feynman, and by his metric, we understand a bit about physics, less about chemistry, and almost nothing about biology. When we fully understand a phenomenon, we can specify its entire sequence of events, causes, and effects so completely that it is possible to fully simulate it, with all its internal mechanisms intact. Achieving that level of understanding is rare. It is commensurate with constructing a full design for a machine that could serve as a stand-in for the thing being studied. To understand a phenomenon sufficiently to fully simulate it is to understand it computationally. 'Computation' does not refer to computers per se. Rather, it refers to the underlying principles and methods that make them work. As Turing Award recipient Edsger Dijkstra said, computational science 'is no more about computers than astronomy is about telescopes.' Computational science is the study of the hidden rules underlying complex phenomena from physics to psychology. Computational neuroscience, then, has the aim of understanding brains sufficiently well to be able to simulate their functions, thereby subsuming the twin goals of science and engineering: deeply understanding the inner workings of our brains, and being able to construct simulacra of them. As simple robots today substitute for human physical abilities, in settings from factories to hospitals, so brain engineering will construct stand-ins for our mental abilities, and possibly even enable us to fix our brains when they break.

研究动机与目标

• 将计算模拟确立为理解脑功能的黄金标准。
• 阐明神经科学中的“计算”指的是底层原理，而非数字计算机。
• 通过模拟统一脑研究中的科学理解与工程应用。
• 论证唯有能够构建类脑系统，才能声称真正理解。
• 将脑工程定位为下一个前沿，类比于物理系统中的机器人学。

提出的方法

• 采用理查德·费曼的原理：‘若我无法构建它，便不代表我理解它。’
• 将‘计算’定义为对复杂系统隐藏规则的研究，而非计算机技术。
• 使用模拟作为测试和验证神经功能模型的手段。
• 以完整的因果和机制保真度建模脑系统，以实现完全模拟。
• 将计算科学与物理学、天文学等其他学科进行类比。
• 将脑工程定义为构建心理能力功能仿真的过程。

实验结果

研究问题

• RQ1 什么标准可定义对脑等生物系统的真实理解？
• RQ2 计算模拟如何作为科学理解的基准？
• RQ3 神经科学中的‘计算’与数字计算有何本质区别？
• RQ4 脑研究中科学洞察力与工程能力之间存在何种关系？
• RQ5 模拟脑功能如何可能带来修复或增强心理能力的实际应用？

主要发现

• 只有当一个现象的完整因果与效应序列能够被模拟时，才可实现真正理解。
• 能够对系统进行计算模拟，等同于理解其底层机制。
• 计算神经科学旨在构建脑功能的功能仿体，融合科学与工程。
• 以模拟作为验证的原则，对脑与物理系统均适用。
• 脑工程有朝一日可能产生可复制或扩展人类心理能力的类人工系统。
• 该方法借鉴了物理学与计算机科学的基础思想，强调原理而非技术。