- 博客(5)
- 资源 (5)
- 收藏
- 关注
RSS订阅原创 Culture model —— Model thinking lecture note (7)
Why cultures are different? Assume that we have a n*n square. Each block represent a person. Assumption 1 : we have several features to coordinate each feature has a trait. a trait looks like {1,2,2...
2018-07-27 10:26:32
254
原创 图解机器学习笔记（五）
L1约束 为了加快学习的速度，我们可以通过将某些学习参数置零达到这个效果，例如，我们可以使用L1约束。 L1约束即： 求解： 因为在这里的绝对值函数的缘故，所以不能把微分项置0，但是通过近似操作，可以得到： 最后解法： ...
2018-07-25 11:11:45
223
原创 lyaponuv function —— Model thinking lecture note (6)
If we can set up a lyaponuv function for the model, then this system means it can go to equilibrium. F(x) is a lyaponuv function. Assumption1: F(x) has a maximum value. Assumption2:if x(t+1) != x(...
2018-07-24 19:48:31
246
原创 CNN review: AlexNet;ResNet;Faster R-CNN;
The input of CNN Some rgb image like 28*28*3 array( the resolution is 28*28 while *3 means r g b three chanel). The output of CNN The probability of several classes like (dog 10% cat 1% others ...)...
2018-07-18 11:26:26
674
原创 Model thinking lecture note (5)
In this part, i will introduce how to solve the problem. (1) Perspective Perspective means how you encode the question. you can explain the question in many ways, but how you explain the question...
2018-07-08 22:24:11
168
正交匹配追踪(OMP)的matlab程序
2018-11-01
db小波去噪后信噪比输出matlab程序
2018-11-01
图解机器学习
2018-05-08
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人