Digest #2019.10.14 – The Ultimate Learning Machine

The Ultimate Learning Machine – Babies… This article dives into how machines are trained and their motivation for learning compared to a human baby. Guess what… babies learn faster, require less data and have a power that machines don’t – curiosity! The AI lab at UC Berkeley is trying to innovate with AI to mimic a child’s curiosity and active learning. “AIs that are motivated by curiosity are more robust and resilient learners than those that are just motivated by immediate rewards.” The test program is cleverly called MESS, short for Model-Building, Exploratory, Social Learning System.

What’s your ML framework? – TensorFlow? PyTorch? Both? An interesting comparison between two of the most popular machine learning heavyweights. The article suggests PyTorch being more prevalent in research and TensorFlow dominating the industry side of Machine Learning. While the reality is not as black and white, the article with the visual data graphs is a good read.

PyTorch 1.3 is here! – And, named tensors, better mobile performance and quantization experiments are live! “The 1.3 release of PyTorch brings significant new features, including experimental support for mobile device deployment, eager mode quantization at 8-bit integer, and the ability to name tensors.” Check out the official announcement for all the exciting details.

Machine Learning for Autonomous Vehicles – Intriguing article about opensource machine learning for autonomous vehicles using Kubeflow. Interesting points on making vehicles autonomous, deep learning services, edge computing and training in the cloud. Ahh, hands-free driving, we’re getting there – fingers crossed!