This Google Experiment Destroyed Some of the Core Assumptions Behind Representation Learning

mediumThis post was originally published by Jesus Rodriguez at Medium [AI]

Two years ago, Google Research published a paper that challenge some of the common assumptions about disentangled representations.

 
 
Source: https://www.digitalvidya.com/blog/representation-learning-101-get-ready-set-go/

Building knowledge in high dimensional datasets is one of the fundamental challenges of modern deep learning applications. As humans, we are extremely proficient of reasoning in a small number of dimensions but information represented in a large number of dimensions results mostly incomprehensible. One ability of humans cognition that proves helpful when understanding large dimensional datasets is our ability to decompose the world in smaller and somewhat disconnected pieces of knowledge. In the context of deep learning, the equivalent to that skill is known as disentangled representations. Two years ago, artificial intelligence(AI) researchers from Google published a paper that challenges the traditional understanding of disentangled representations.

Disentangled Representations and Representation Learning

 
Image Credit: Google
 
Image Credit: Google

The Google Experiment

 
Image Credit: Google
 
Image Credit: Google

Spread the word

This post was originally published by Jesus Rodriguez at Medium [AI]

Related posts