Stop Doing Fragile Research

[unable to retrieve full-text content]If you develop methods for data analysis, you might only be conducting gentle tests of your method on idealized data. This leads to “fragile research,” which breaks when released into the wild. Here, I share 3 ways to make your methods robust.
Original Post: Stop Doing Fragile Research

You have created your first Linear Regression Model. Have you validated the assumptions?

[unable to retrieve full-text content]Linear Regression is an excellent starting point for Machine Learning, but it is a common mistake to focus just on the p-values and R-Squared values while determining validity of model. Here we examine the underlying assumptions of a Linear Regression, which need to be validated before applying the model.
Original Post: You have created your first Linear Regression Model. Have you validated the assumptions?

The amazing predictive power of conditional probability in Bayes Nets

[unable to retrieve full-text content]This article explains how Bayes Nets gain remarkable predictive power by their use of conditional probability. This adds to several other salient strengths, making them a preeminent method for prediction and understanding variables’ effects.
Original Post: The amazing predictive power of conditional probability in Bayes Nets

How Bayesian Networks Are Superior in Understanding Effects of Variables

[unable to retrieve full-text content]Bayes Nets have remarkable properties that make them better than many traditional methods in determining variables’ effects. This article explains the principle advantages.
Original Post: How Bayesian Networks Are Superior in Understanding Effects of Variables

Choosing an Open Source Machine Learning Library: TensorFlow, Theano, Torch, scikit-learn, Caffe

[unable to retrieve full-text content]Open Source is the heart of innovation and rapid evolution of technologies, these days. Here we discuss how to choose open source machine learning tools for different use cases.
Original Post: Choosing an Open Source Machine Learning Library: TensorFlow, Theano, Torch, scikit-learn, Caffe