Intuition Machine - Issue #57

Experience flexible research and accelerated production with Facebook’s ecosystem of open source, state-of-the-art AI developer tools.

Today, Facebook AI Research (FAIR) open sourced ELF OpenGo, an AI bot that has defeated world champion professional Go players, based on our existing ELF platform for Reinforcement Learning Research. 

A broad ML benchmark suite for measuring performance of ML software frameworks, ML hardware accelerators, and ML cloud platforms.

New Deep Learning Techniques (Schedule) - IPAM

The brain is complex; in humans it consists of about 100 billion neurons, making on the order of 100 trillion connections. It is often…

The BAIR Blog

You can enter multiple addresses separated by commas to send the article to a group; to send to recipients individually, enter just one address at a time.

Recent discussions about the “advent of robots” have some rather unusual features. The threat of robots replacing humans is seen as something truly novel possibly changing our civilization and way of life. But in reality this is nothing new. Introduction of machinery to replace repetitive (or even more creative) labor has been applied on a significant scale since  the beginning of the Industrial Revolution. Robots are not different from any other machine.

Artificial Intelligence (AI) is the mantra of the current era. The phrase is intoned by technologists, academicians, journalists and venture capitalists alike. As with many phrases that cross over…

A  movie where privacy is the ultimate crime.

If you combine two or three metals together, you will get an alloy that usually looks and acts like a metal, with its atoms arranged in rigid geometric patterns.

Tax forms filed by OpenAI provide insight into the enormous salaries and bonuses paid to artificial intelligence specialists across the world.

Over the last six months, a powerful new neural network playbook has come together for Natural Language Processing. The new approach can be summarised as a simple four-step formula: embed, encode, attend, predict. This post explains the components of this new approach, and shows how they’re put together in two recent systems.

We recently published an article announcing five papers on deep neuroevolution, including the discovery that genetic algorithms can solve deep reinforcement learning problems as well as popular alternatives, such as deep Q-learning and policy gradients. 

How I discovered a new deep learning paper on Twitter, replicated its results, and submitted an open-source implementation I got accepted to the fast.ai International Fellowship at the end of…

Introduction Short intro to Python extension objects in C/C++ Zero-copy PyTorch Tensor to Numpy and vice-versa Tensor Storage Shared Memory DLPack: a hope for the Deep Learning frameworks Babel Introduction This post is a tour around the PyTorch codebase, it is meant to be a guide for the architectural design of

The virtual personal assistant is romanticized in utopian portrayals of the future from The Jetsons to Star Trek. It’s the cultured, disembodied voice at humani…

UI Engineering and Design consultant, specializing in React and React performance. Previously at Facebook and Storehouse. Winner of the first Node.js Knockout with Swarmation.com.

by Nicolas Papernot and Ian Goodfellow

A reincarnation of one of the oldest ideas in artificial intelligence could finally make it possible to truly converse with our computers. And Facebook has a chance to make it happen first.

When you teach a neural network to see, you also teach it to draw.

by Nicolas Papernot and Ian Goodfellow