AI labs – Learning unsupervised learning through Robotics

We are launching an AI lab. The goal is to learn unsupervised learning through Robotics (Cobots)
Long seen as a poor cousin to supervised learning –  with Variational autoencoders, Reinforcement learning and  Generative-Adversarial networks , unsupervised learning techniques have moved beyond the limitations of autoencoders.
From Oct 2018 to March 2019 , we are running a pilot.

Read Full Story

Paper: A Differentiable Physics Engine for Deep Learning in Robotics

Abstract:
One of the most important fields in robotics is the optimization of controllers. Currently, robots are often treated as a black box in this optimization process, which is the reason why derivative-free optimization methods such as evolutionary algorithms or reinforcement learning are omnipresent.

Read Full Story

Learning to Assemble and to Generalize from Self-Supervised Disassembly

Posted by Kevin Zakka, Research Intern and Andy Zeng, Research Scientist, Robotics at Google Our physical world is full of different shapes, and learning how they are all interconnected is a natural part of interacting with our surroundings — for example, we understand that coat hangers hook onto clothing racks, power plugs insert into wall outlets, and USB cables fit into USB sockets.

Read Full Story