Digest #2019.10.07 – Machine Learning in Space? Agriculture?

How NASA uses Machine Learning – If you think the Earth is the only planet with Machine Learning you were wrong. The Mars Rover learning its path and environment at Mars? The healthcare needs for future astronauts? Planet exploration and discovery? Robotic astronaut!? All of these are questions NASA is experimenting and researching with Machine Learning. Check out the intriguing full article for details.

TensorFlow 2.0 is here! – For those of you who haven’t already heard, TensorFlow 2.0 is official! And, with TF 2.0 Keras is also official [Now the official high-level API for TensorFlow]. TensorFlow can now be run like Python code, yes, you can see the values of variables with print()! The APIs have been refined and there’s an upgrade script for a seamless transition from 1.x to 2.0. Oh, and did I mention backwards compatibility? Check out the official announcement for all exciting changes.

The benefits of AI and ML – With all the skepticism and fear of AI and ML, here’s a refreshing newspaper article that tells people not to be afraid. The robots aren’t coming for us. Not necessarily, not yet… 

Agriculture as a tough Machine Learning problem – You know Machine Learning is legit when a centuries-old tractor manufacturer starts classifying(no pun intended) itself as a software company. John Deere Labs believes it is solving one of the toughest Machine Learning problems, classification of grains! The article explores the possibility of classifying good and bad grains of corn and making harvesting adjustments accordingly. There’s a lot of deep learning and classification challenges to solve that can revolutionize the harvest. How exciting!