Luc Frachon – Medium

Pinned

5 Reasons to Use Linear Regression over Gradient Boosted Trees (or at least give it a go)

Introduction In the age of deep learning and trillion-parameter models, it is easy to ignore good old linear regression that’s been around for over two centuries. Of course, linear regression alone is not robust enough for many modern AI applications. Even when considering only tabular data, plenty of models will outperform…

Machine Learning

5 min read

5 Reasons to Use Linear Regression over Gradient Boosted Trees (or at least give it a go)

Machine Learning

5 min read

1 day ago

Anomaly Detection Using a Variational Autoencoder, Part II

Part II: Implementation Anomaly detection is one of the most widespread use cases for unsupervised machine learning, especially in industrial applications. In Part I, we motivated the use of variational autoencoders for this task and provided intuitive explanations of how to do it with references to dogs and sheep. …

Anomaly Detection

12 min read

Anomaly Detection

12 min read

Nov 16

Anomaly Detection Using a Variational Autoencoder, Part I

Part I: Motivation and Intuition Anomaly detection is one of the most widespread use cases for unsupervised machine learning, especially in industrial applications. In this post, I will explain why and how to leverage a variational autoencoder’s learning abilities to detect anomalies, with intuitive explanations involving dogs and sheep (yes, really). …

Machine Learning

7 min read

Anomaly Detection Using a Variational Autoencoder, Part I

Machine Learning

7 min read