Intuition Machine - Issue #51

Centralized platforms have been dominant for so long that many people have forgotten there is a better way to build internet services.

Strategy for Disruptive Artificial Intelligence. Visit https://deeplearningplaybook.com for more information. 

This is an illustration of a multi-compartment neural network model for deep learning. Left: Reconstruction of pyramidal neurons from mouse primary visual

This tutorial walks through a synchronous single-thread single-GPU (read malnourished) game-agnostic implementation of the recent AlphaGo Zero paper by DeepMind. It’s a beautiful piece of work that trains an agent for the game of Go through pure self-play without any human knowledge except the rules of the game. 

Stream 36 - Attention Is All You Need, with Ashish Vaswani and Jakob Uszkoreit by NLP Highlights from desktop or your mobile device

We’ve co-authored a paper that forecasts how malicious actors could misuse AI technology, and potential ways we can prevent and mitigate these threats. 

Abstract: Intelligent creatures can explore their environments and learn useful skills without supervision. In this paper, we propose a method for learning useful skills without a reward function. We maximize an information theoretic objective using a maximum entropy policy.

An outsider tour of reinforcement learning, Part 6. Policy gradient doesn’t have gradients.

⊕

Online demo

Numerical optimization is at the core of much of machine learning. In this post, we derive the L-BFGS algorithm, commonly used in batch machine learning applications.

Deep neural networks can be trained with relatively modest amounts of information and then successfully be applied to large quantities of unstructured data. …

Juan Carrasquilla gave himself a crash course on machine learning and found a new way of approaching condensed-matter theory.