MIT researchers have developed a computer system that independently adds realistic sounds to silent videos. Although the technology is nascent, it's a step toward automating sound effects for movies. In a series of videos of drumsticks striking things -- including sidewalks, grass and metal surfaces -- the computer learned to pair a fitting sound effect, such as the sound of a drumstick hitting a piece of wood or of rustling leaves. The findings are an example of the power of deep learning, a type of artificial intelligence whose application is trendy in tech circles. With deep learning, a computer system learns to recognize patterns in huge piles of data and applies what it learns in useful ways.

Today at WWDC, Apple brought machine learning to Photos to help you find, discover and share your images in a more intuitive way than ever before. The features borrow some of the best features from Google Photos, like re-surfacing memorable events, creating albums based on events, people and places, and using deep learning to help find images in a more intuitive way. The new algorithm uses advanced computer vision, a group of deep learning techniques that brings facial recognition to the iPhone. Now, you can find all of the most important people, places and things in your life in with automatically sorted albums. It's essentially facial recognition that works on places and objects as well.

For robots to navigate the world, they need to be able to make reasonable assumptions about their surroundings and what might happen during a sequence of events. One way that humans come to learn these things is through sound. For infants, poking and prodding objects is not just fun; some studies suggest that it's actually how they develop an intuitive theory of physics. Could it be that we can get machines to learn the same way? Researchers from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) have demonstrated an algorithm that has effectively learned how to predict sound: When shown a silent video clip of an object being hit, the algorithm can produce a sound for the hit that is realistic enough to fool human viewers.

Three actors in glittery metallic costumes stand in an office, spouting gibberish at one another: the outcome of an experiment that couldn't be any more predictable. Earlier this week, UK filmmaker Oscar Sharp released Sunspring, an experimental film based on a screenplay developed by an artificial intelligence. The project saw Mr Sharp and technologist Ross Goodwin feeding an artificial long short-term memory neural network (LSTM) an array of science fiction film scripts. Mr Sharp and Mr Goodwin then programmed the LSTM to process the scripts, using cinematic writing conventions to produce a screenplay of its own. Mr Sharp then spent 48 hours developing the script into a film, before releasing it on the Ars Technica Videos YouTube page on June 9.

Using machine learning, researchers from MIT have developed a system that produces sound effects that are so realistic they even fool human listeners. The new algorithm, developed by researchers from MIT's Computer Science and Artificial Intelligence Laboratory, can predict the precise acoustical qualities of a sound, and then simulate it in an extremely realistic way. When analyzing a silent video clip, such as an object being hit by a drumstick, the system can produce a sound for the hit that's realistic enough to fool human listeners. To make it work, PhD student Andrew Owens and his team applied a technique known as "deep learning" that enables computers to pick out important patterns buried in massive amounts of raw data completely autonomously. Over the course of several months, the researchers recorded about 1,000 videos of an estimated 46,000 sounds that represented an array of objects being hit, scraped, and prodded by a drumstick.

In the movie Transcendence, Johnny Depp plays Dr Will Caster, a researcher in artificial intelligence at Berkeley trying to build a sentient computer. Stuart Russell is Will Caster's real life equivalent. He works on artificial intelligence at the University of California at Berkeley, and is co-author of the definitive textbook on AI. He has also been very vocal about the risks of research in AI succeeding. Earlier this year, Google's DeepMind taught a computer program to play a wide variety of Atari video games at a superhuman level in a matter of hours.

Here are Dockerfiles to get you up and running with a fully functional deep learning machine. It contains all the popular deep learning frameworks with CPU and GPU support (CUDA and cuDNN included). The CPU version should work on Linux, Windows and OS X. If you are not familiar with Docker, but would still like an all-in-one solution, start here: What is Docker?. GPU Version Only: Install Nvidia drivers on your machine either from Nvidia directly or follow the instructions here. Note that you don't have to install CUDA or cuDNN.

Android N notifications are getting a new look to help provide a better user experience. They now have a fresh look, improved for custom views, and expanded functionality in the forms of Direct Reply. The default look and feel of notifications has changed, and now the fields around the notifications have been collapsed into a new header row with the app's icon and name anchoring the notification, according to Android developer advocate Ian Lake on the Android Developers Blog. This makes the title, text and large icon have a lot of space, and now the notifications are slightly larger and easier to read. Notification actions have also received a redesign and are now in a visually separate bar below the notification, according to the blog.

In this work, we consider compressed sensing reconstruction from $M$ measurements of $K$-sparse structured signals which do not possess a writable correlation model. Assuming that a generative statistical model, such as a Boltzmann machine, can be trained in an unsupervised manner on example signals, we demonstrate how this signal model can be used within a Bayesian framework of signal reconstruction. By deriving a message-passing inference for general distribution restricted Boltzmann machines, we are able to integrate these inferred signal models into approximate message passing for compressed sensing reconstruction. Finally, we show for the MNIST dataset that this approach can be very effective, even for $M < K$.