Monthly Archives: January 2025

I woke up one morning with a mild cold and headache. The headache wasn’t too bad, but it was bad enough that I didn’t want to work on difficult topics like PyTorch neural systems or simulated quantum computing. So I … Continue reading →

Suppose you have a regression problem with three predictor variables, x0, x1, x2. Standard linear regression creates a prediction equation like: y = (w0 * x0) + (w1 * x1) + (w2 * x2) + b where w0, w1, w2 … Continue reading →

Gradient boosting regression (GBR) is a common machine learning technique to predict a single numeric value. The technique combines many small decision trees (typically 100 or 200) to produce a final prediction. One morning before work, I figured I’d put … Continue reading →

I wrote an article titled “Random Forest Regression and Bagging Regression Using C#” in the January 2025 edition of Microsoft Visual Studio Magazine. See https://visualstudiomagazine.com/Articles/2025/01/02/Random-Forest-Regression-and-Bagging-Regression-Using-CSharp.aspx. A machine learning random forest regression system predicts a single numeric value. A random forest … Continue reading →

A regression problem is one where the goal is to predict a single numeric value. Common machine learning regression techniques are linear regression, k-nearest neighbors, kernel ridge, Gaussian process, decision tree, AdaBoost (includes bagging), gradient boost, and kernel ridge regression … Continue reading →

Gaussian process regression (GPR) is a powerful machine learning technique to predict a single numeric value. When GPR works, it often works very well. The major disadvantages of GPR are: 1.) it uses matrix inversion, which means GPR is not … Continue reading →