Editor's Pick
Robots that Learn
Elon Musk's OpenAI created a robotics system, trained entirely in simulation and deployed on a physical robot, which can learn a new task after seeing it done once.
openai.com
News & Articles
$100M For A New Quebec Research Cluster in AI
The Quebec government today announced the creation of a steering committee to develop a strategic plan for the creation of a $100-million research cluster in artificial intelligence based in Montreal.
umontreal.ca
Nvidia's monstrous Volta GPU appears, packed with 21 billion transistors and 5,120 cores
From recognizing speech to training virtual personal assistants to converse naturally; from detecting lanes on the road to teaching autonomous cars to drive; data scientists are taking on increasingly complex challenges with AI. Solving these kinds of problems requires training exponentially more complex deep learning models in a practical amount of time.
nvidia.com
Open Source Deep Learning Frameworks and Visual Analytics
This article explains why Deep Learning is a game changer in analytics, when to use it, and how Visual Analytics allows business analysts to leverage the analytic models built by a (citizen) data scientist.
datasciencecentral.com
Call for Papers
CoRL2017: 1st Annual Conference on Robot Learning
Paper submission deadline: June 28, 2017
robot-learning.org
Papers & Tutorials
Pix2Pix: Generic Image-to-Image Translation Using Conditional Adversarial Networks
This tutorial will guide you on how to use the pix2pix software for learning image transformation functions between parallel datasets of corresponding image pairs.
github.io
Learning Important Features Through Propagating Activation Differences
"DeepLIFT compares the activation of each neuron to its 'reference activation' and assigns contribution scores according to the difference. By optionally giving separate consideration to positive and negative contributions, DeepLIFT can also reveal dependencies which are missed by other approaches."
arxiv.org
Video
How to Simulate a Self-Driving Car
Using Keras to train a convolutional neural network on images from the car's cameras as well as steering angles from human driving. Using just those 2 data points, it'll be able to drive itself on any road.
youtube.com
