The Facebook Ads team have launched a field guide to machine learning video series. The six videos include work practices and tips at applying machine learning to real world problems.
Google have launched The Lever, a blog that presents the best practices in applying machine learning, including posts about how to approach adding machine learning to your product, and posts about how to gather or acquire data.
This Smashing Magazine tutorial demonstrates how to build a machine learning model that can predict which room you are in based on the signal strength of wifi networks around you. The system requires a map of the networks to be recorded and a model trained using the recorded data. Such a system could be useful to passively trigger IOT devices based on which room your phone is in, without the need for active notification methods like bluetooth beacons.
Researchers have used machine learning and AI to identify a brain signal model that characterises Fibromyalgia, despite the some doctors believing that Fibromyalgia does not exist. The signature model can be used for assessing therapeutic mechanisms and predicting treatment response at an individual patient level.
U.S. creative agency RedPepper has built a Raspberry Pi powered robot arm with AI image recognition which can successfully find Waldo. The system uses OpenCV to extract the faces from the image and Google Auto ML Vision cloud service to match the faces to Waldo model. No word as of yet on whether Waldo's British older brother Wally has escaped the robot overlords :-p
This is a great demonstration post by Joel Simon on using genetic algorithms to optimise a school floorplan. He demonstrates an optimised layout for traffic flow, fire escape routes and access to windows among other things.
Reseachers at UCLA have developed an all optical neural network system, which uses specially etched plates to diffract incoming light to perform calculations. A number of plates are layered together to create teh Diffractive Deep Neural Network (D2NN) which can perform calculations at the speed of light.
Google have released two new Edge TPU hardware implementations available for its Tensorflow Processing Unit. The machine learning chips can be embedded on dev board or bundled into a USB stick device. Both devices are due to be available in the Autumn.
Two PhD students at Cambridge University have kitted out a Renault Twizzy with a front facing camera and used it to teach reinforment algorithm in 20 minutes to drive along a path without veering off.
R2D3 have created a fantastic data visualisation describing how data models can be tuned to avoid over-emphases on biases of variances during the training phases.