非线性回归和偏微分

线性回归和非线性回归是统计学里的概念。机器学习(百度百科)

非线性回归(百科) | Machine Learning [Python] – Non-linear Regression (geekering.com) | NonlinearRegression.dvi (toronto.edu)

y(x) = w₀ + w₁ * x 是一元线性回归。w代表weight(权重), 只有x一个元。

y(x) = w₀ + w₁x₁ + w₂x₂ + … + w_nx_n 是多元线性回归。有n个元和n+1个权重。

Sometimes linear models are not sufficient to capture the real-world phenomena, and thus nonlinear models are necessary. In linear regression, we have f(x) = Wx + b; the parameters W and b must be fit to data. What nonlinear function do we choose? In principle, f(x) could be anything: it could involve linear functions, sines and cosines, summations, and so on. However, In many situations, we do not know much about the underlying nature of the process being modeled, or else modeling it precisely is too difficult. In these cases, we typically turn to a few models in machine learning that are widely-used and quite effective for many problems. These methods include basis function regression (including Radial Basis Functions), Artificial Neural Networks, and k-Nearest Neighbors.

f(x) = ax³ + bx² + cx + d 是非线性回归；我是非穷人。

Partial derivative

  即把y看作常数。导数(百科)

 

深度学习利器之自动微分(1) - 罗西的思考 - 博客园 (cnblogs.com)

深度学习利器之自动微分(2) - 罗西的思考 - 博客园 (cnblogs.com)

python写的多项式符号乘法 - Fun_with_Words - 博客园 (cnblogs.com)

D:\>poly.py
(x - 1) * (x^2 + x + 1) = x^3 - 1

条件概率 - Fun_with_Words - 博客园 (cnblogs.com) 概率只研究现象，不解释原因。