Digests » 148

this week's favorite

Sync GAN Art to Music with a Few Lines of Python Code

Generative art has made significant strides in the past few years. One need only look at the countless artworks posted across the Internet (e.g. here and here) to realize that machines have arguably equaled man in terms of artistic proficiency.

Algorithm for Training Neural Nets with Bayesian Linear Regression

The idea behind this is that when using ReLU activation functions the activation is either linear with the input (if it’s activated) or zero (if it’s not activated). So the activated neurons can be updated using Bayesian Linear Regression, and we leave the unactivated ones alone.

Why machine learning struggles with causality

When you look at the following short video sequence, you can make inferences about causal relations between different elements. For instance, you can see the bat and the baseball player’s arm moving in unison, but you also know that it is the player’s arm that is causing the bat’s movement and not the other way around. You also don’t need to be told that the bat is causing the sudden change in the ball’s direction.

Introduction to Anomaly Detection with Auto-Encoder on time series

An auto encoder is a neural network that learns a representation of data in a smaller dimension. In concrete terms, it is given a vector and is expected to return the same vector, in other words, it rebuilds its entry.

SpeechBrain: A PyTorch Speech Toolkit

SpeechBrain is an open-source and all-in-one speech toolkit. It is designed to be simple, extremely flexible, and user-friendly. State-of-the-art performance are obtained in various domains.