This Neural Net Hallucinates Sheep

How to play Quidditch using the TensorFlow Object Detection API

“Cracking” Morse code with RNNs

Spoiler alert: Morse code doesn’t really need cracking. Its useful because messages can be sent using this code with minimal equipment, and I say it doesn’t need cracking because the code is well known and what the combinations of dots and dashes stand for is no secret. But, in theory, it is a substitution cipher — where each letter of the alphabet (and each digit) has some representation using dots and dashes, as illustrated below.

Learning by playing

Our new paper proposes a new learning paradigm called ‘Scheduled Auxiliary Control (SAC-X)’ which seeks to overcome the issue of exploration in control tasks. SAC-X is based on the idea that to learn complex tasks from scratch, an agent has to learn to explore and master a set of basic skills first. Just as a baby must develop coordination and balance before she crawls or walks—providing an agent with internal (auxiliary) goals corresponding to simple skills increases the chance it can understand and perform more complicated tasks.