All Stories
GAN optimality proof
The so-called Generative Adversarial Networks have been with us since 2014, producing amazing results lately. Today I am bringing the proof of why they work. That is, a proof that...
Neural Architecture Search (Part 2)
In the previous article we discussed how we can use reinforcement learning to design simple architectures like some types of convolutional neural networks. Today I am bringing to you the...
Neural Architecture Search
Today we are going to dive into an idea that some may fear, and others may praise: AI training itself. Well, in reality the idea is a bit different from...