The paper "Model Compression" is available here: https://www.cs.cornell.edu/ We also thank Experiment for sponsoring our series. Subscribe if you would like to see more of these! - http://www.youtube.com/subscription_c... The thumbnail background image was created by John Lord - https://flic.kr/p/nVUaB

A supervised machine learning model aims to learn a function f(x) y from a list of training pairs (x1,y1), (x2,y2), … for which data are recorded (Fig 1B). One typical application in biology is to predict the viability of a cancer cell line when exposed to a chosen drug (Menden et al, 2013; Eduati et al, 2015). The input features (x) would capture somatic sequence variants of the cell line, chemical make?up of the drug and its concentration, which together with the measured viability (output label y) can be used to train a support vector machine, a random forest classifier or a related method (functional relationship f). Given a new cell line (unlabelled data sample x*) in the future, the learnt function predicts its survival (output label y*) by calculating f(x*), even if f resembles more of a black box, and its inner workings of why particular mutation combinations influence cell growth are not easily interpreted. Both regression (where y is a real number) and classification (where y is a categorical class label) can be viewed in this way. As a counterpart, unsupervised machine learning approaches aim to discover patterns from the data samples x themselves, without the need for output labels y. Methods such as clustering, principal component analysis and outlier detection are typical examples of unsupervised models applied to biological data. The inputs x, calculated from the raw data, represent what the model "sees about the world", and their choice is highly problem?specific (Fig 1C). Deriving most informative features is essential for performance, but the process can be labour?intensive

All architectures are compatible with both TensorFlow and Theano, and upon instantiation the models will be built according to the image dimension ordering set in your Keras configuration file at /.keras/keras.json. For instance, if you have set image_dim_ordering tf, then any model loaded from this repository will get built according to the TensorFlow dimension ordering convention, "Width-Height-Depth". Weights can be automatically loaded upon instantiation (weights'imagenet' argument in model constructor). Weights are automatically downloaded if necessary, and cached locally in /.keras/models/. Note that using these models requires the latest version of Keras (from the Github repo, not PyPI). Additionally, don't forget to cite Keras if you use these models.

With new neural network architectures popping up every now and then, it's hard to keep track of them all. Knowing all the abbreviations being thrown around (DCIGN, BiLSTM, DCGAN, anyone?) can be a bit overwhelming at first. So I decided to compose a cheat sheet containing many of those architectures. Most of these are neural networks, some are completely different beasts. Though all of these architectures are presented as novel and unique, when I drew the node structures… their underlying relations started to make more sense. One problem with drawing them as node maps: it doesn't really show how they're used. For example, variational autoencoders (VAE) may look just like autoencoders (AE), but the training process is actually quite different. The use-cases for trained networks differ even more, because VAEs are generators, where you insert noise to get a new sample. AEs, simply map whatever they get as input to the closest training sample they "remember". I should add that this overview is in no way clarifying how each of the different node types work internally (but that's a topic for another day).

IBM has launched three Power8 Linux servers designed to accelerate artificial intelligence, deep learning, and advanced analytics applications. The new systems tap the Nvidia NVLink technology to move data five times faster than any competing platform, said Stefanie Chiras, an IBM vice president, in an interview with VentureBeat. These systems and their operating systems are part of a larger business group that generates about 2 billion a quarter for IBM. And the A.I. markets they're going after have exploded in the past couple of years. IBM claims that the combination of Power8 processors and Linux software results in systems that deliver 80 percent more performance per dollar than the latest x86-based (Intel or AMD) servers.

Training neural networks involves quite a few tricky bits. We try to make everything clear and easy to understand, to get you training your neural networks as quickly as possible. Theano allows us to write relatively concise code that follows the structure of the underlying maths. We will train a network to classify digits. More precisely, we want a network that when presented with an image of a hand-written digit will tell us what digit it is ('0', '1', ..., '9').

Back in 2012, a team at Google built a state-of-the-art artificial intelligence network and fed it ten million randomly selected images from YouTube. The computer churned through them, and announced that it kept finding these strange things with furry faces. It had, in other words, discovered cats. Artificial intelligence has, all of a sudden, become the next big thing again. It is not so much sweeping across our world as seeping into it, with a combination of enormous computing power and the latest'deep learning' techniques promising to give us better medical diagnoses, better devices, better recipes and better lives.

Drones and robots are getting computer vision with higher-resolution cameras and artificial intelligence to recognize objects and images. Many are made with developer boards like Nvidia's Jetson TX1, which provides the smarts for auto-navigation and collision avoidance. TX1 has the horsepower to process live image feeds, and software tools to instantly analyze and provide context to visuals. The TX1 is now a lot faster and better equipped to handle AI and image processing. Nvidia's new Jetpack 2.3 software tools for TX1, announced on Tuesday, are a major update that doubles the deep-learning performance of the board.

June 1, 2016, The Internet of Things topped the target list for developers working with artificial intelligence across a wide spectrum of technologies including machine learning, neural networks, deep learning, and pattern recognition, according to Evans Data's just released Global Development Survey. While targets for these technologies remain fragmented, IoT was the top target for all of them and in most cases the only target with a double digit response. Non-computer related professional, scientific and technical services was cited second as a target for the above disciplines, and was first in the category of Natural Language Processing. "All the related disciplines that are commonly lumped together as artificial intelligence are being stimulated by the burgeoning growth of Internet of Things," said Janel Garvin, CEO of Evans Data. "These technologies are being incorporated very rapidly into the design and development process across a host of industries, and types of applications, but it's IoT that is the strongest driver."

A new personalized search engine helps you explore what you would look like with brown hair, curly hair or in a different time period. Ira Kemelmacher-Shlizerman, a computer vision researcher at University of Washington, developed the image recognition software using a TITAN X GPU and the cuDNN-accelerated Caffe deep learning framework to train the models and for inference. Ira presented her paper at this week's SIGGRAPH 2016 and the search engine will be publicly available later this year. Dreambit is also able to predict what a child might look like when they are forty years old or with red hair, black hair, or even a shaved head. "It's hard to recognize someone by just looking at a face, because we as humans are so biased towards hairstyles and hair colors," said Kemelmacher-Shlizerman. "With missing children, people often dye their hair or change the style so age-progressing just their face isn't enough. This is a first step in trying to imagine how a missing person's appearance might change over time."