nico.ash + machine-learning   13

Differentiable Plasticity: A New Method for Learning to Learn | Uber Engineering Blog
Uber AI Labs has developed a new method called differentiable plasticity that lets us train the behavior of plastic connections through gradient descent so that they can help previously-trained networks adapt to future conditions. While evolving such plastic neural networks is a longstanding area of research in evolutionary computation, to our knowledge the work introduced here is the first to show it is possible to optimize plasticity itself through gradient descent. Because gradient-based methods underlie many of the recent spectacular breakthroughs in artificial intelligence (including image recognition, machine translation, Atari video games, and Go playing), making plastic networks amenable to gradient descent training may dramatically expand the power of both approaches.
ai  learning  machine-learning  ml 
april 2018 by nico.ash
Show HN: Sorting Two Metric Tons of Lego | Hacker News
A thread about a self-built machine that sort bulk lego parts into bins using machine learning.
machine-learning  lego  opencv 
april 2017 by nico.ash
Suddenly, a leopard print sofa appears
A blog post about the pitfalls of training and using machine learning for content recognition in images.
machine-learning  ai 
june 2015 by nico.ash

Copy this bookmark:



description:


tags: