Summary of Mathematics in Machine Learning Algorithms
26 Aug 2015This post covers a partial list of parametric machine learning algorithms.
I'm a Master student in Computer Science and Biostatistics(PhD candidate). I love data science and have passion in machine learning.
This post covers a partial list of parametric machine learning algorithms.
Grid search is the most intuitive way of performing hyperparameter optimization. However, it suffers from the curse of dimensionality becuase the number of joint values grows exponentially with the number of hyperparameters. Here is a list of packages, which use better methods of hyperparameter optimization.
Kernel trick is an important concept in machine learning, which is essential to achieve significantly faster computational speed for kernel method based algorithms. (e.g. SVMs, kernel KNN, kernel regression, etc.) In this post, I’d like to introduce in the context of Support Vector Machines 1)why do we need kernel trick, 2)how is it derived step by step, and 3) how does it help to reduce computation complexity.
As a person who starts diving into machine learning field with a statistics background, one thing always confuse me - What 's the difference between statistical modeling and machine learning?