An AI system identified a potential new drug in just 46 days by Charlotte Jee via @techreview

Using general adversarial networks (GANs) and reinforcement learning, researchers at the University of Toronto working with AI startup Insilico Medicine, created 30,000 designs for molecules that target a protein linked with fibrosis (tissue scarring) in 21 days. Six of these molecules were synthesized in the lab, two of those were tested in cells, and the most promising one was tested in mice. The researchers concluded the mouse-tested molecule was potent against the protein and showed “drug-like” qualities. The entire process took just 46 days.

Just wow - thi is some great progress and I just want this to be used against all sorts of illnesses to make some amazing progress. This is definitely a technique that requires further investigation and experimenting so it can be used.

Deep learning is about to get easier by Ben Dickson via @VentureBeat

One problem with deep learning algorithms is that they require vast amounts of data. Fortunately, researchers have found workarounds that will level the playing field.

This is really interesting and anything that helps more people take advantage of AI has got to be a great thing (if it has been tested to make sure you can rely on the answers).

Google’s Duelling Neural Networks Spar to Get Smarter, No Humans Required by Cade Metz via @WIRED

“What an AI cannot create, it does not understand.”

Very interesting and certainly added to my understanding of a few things.