NVIDIA (NVDA) is moving fast in its mission to make AI (artificial intelligence) affordable. The company is expanding its efforts to create more AI researchers and developers through its DLI (Deep Learning Institute). Don't miss the next report. You are now receiving e-mail alerts for new research. A temporary password for your new Market Realist account has been sent to your e-mail address.

So it takes a snippet of speech and then translates the snippet of speech using the voice style of another person. The surprising point though of this research is that its able to encode an internal representation of a speech absent the speaking style. Of course, that sounds like voice to text translation. Now it somehow is able to take out a speech style and transpose it elsewhere. The approach in the paper uses auto-regressive networks, one of those curiously strange thingamajigs that DeepMind seems to be enamored with. It is the same kind of network as WaveNet.

While the current class of deep learning techniques is helping fuel the AI wave, one of the frequently cited drawbacks is that they require a lot of data to work. But how much is enough data? "I would say pretty much any business that has tens or hundreds of thousands of customer interactions has enough scale to start thinking about using these sorts of things," Jeff Dean, a senior fellow at Google, said in an onstage interview at the VB Summit in Berkeley, California. "If you only have 10 examples of something, it's going to be hard to make deep learning work. If you have 100,000 things you care about, records or whatever, that's the kind of scale where you should really start thinking about these kinds of techniques."

Not content to simply transform the worlds of energy, transportation, and space exploration, in 2015, Elon Musk founded OpenAI, a San Fransisco-based artificial intelligence (AI) research company. The non-profits' goal is to further the technology in ways that will benefit humanity as a whole, and over the past two years, they've pushed AI into new territory. Recently, several researchers from OpenAI stepped away from the company to found Embodied Intelligence, a robotics start-up with a more singular focus: propel robotic automation to a higher level. Through their previous work, the founding members of Embodied Intelligence -- former OpenAI researchers Peter Abbeel, Peter Chen, and Rocky Duan and former Microsoft researcher Tianhao Zhang -- explored the potential of robots to mimic complex human action. Now, they are now confident they can use their past experience to improve the type of robots that are currently used in industry and even in the home.

With the major intrusion of technological singularity in our current world, the potential of every AI we create, to perform specific tasks is beginning to explode. Getting to the depth of technological singularity is resulting in developing a technology where machines will replicate their own self. Google in its annual I/O developer conference, disclosed its one of the outlandish yet expected project, named as -- AutoML. This is designed to solve the toughest part of creating an entire AI itself. It majorly revolves around fabricating the deep learning software.

Soumith Chintala is a Researcher at Facebook AI Research. This talk will cover PyTorch, a new deep learning framework that enables new-age A.I. research using dynamic computation graphs. It looks at the architecture of PyTorch and discusses some of the reasons for key decisions in designing it and subsequently look at the resulting improvements in user experience and performance.

Smartphone cameras are pretty incredible things to have in your pocket, and the Pixel 2 does a very good job of making every image look fantastic. But you can't do better than a big, full-frame DSLR – the trouble is, they're not very pocket-friendly. So, if you're fed up of your phone taking washed out, shallow photos, this AI is designed to take your old smartphone pictures and give them DSLR-like quality – even if your smartphone isn't all that snazzy. Known as WESPE (Weakly Supervised Photo Enhancer), the team of data scientists behind the project aim to bring DSLR-like qualities to smartphone cameras. The idea is that, by training a deep learning system using photos of the same scene taken with a phone camera and on a DSLR, it'll learn the difference and automatically make those adjustments on images it's never seen before.

On December 2nd, 1942, a team of scientists led by Enrico Fermi came back from lunch and watched as humanity created the first self-sustaining nuclear reaction inside a pile of bricks and wood underneath a football field at the University of Chicago. Known to history as Chicago Pile-1, it was celebrated in silence with a single bottle of Chianti, for those who were there understood exactly what it meant for humankind, without any need for words. Now, something new has occurred that, again, quietly changed the world forever. Like a whispered word in a foreign language, it was quiet in that you may have heard it, but its full meaning may not have been comprehended. However, it's vital we understand this new language, and what it's increasingly telling us, for the ramifications are set to alter everything we take for granted about the way our globalized economy functions, and the ways in which we as humans exist within it. The language is a new class of machine learning known as deep learning, and the "whispered word" was a computer's use of it to seemingly out of nowhere defeat three-time European Go champion Fan Hui, not once but five times in a row without defeat.

I get bored from doing two things: i) spot-checking optimising parameters of my predictive models and ii) reading about how'black box' machine learning (particularly deep learning) models are and how little we can do to better understand how they learn (or not learn, for example when they take a panda bear as a vulture!). In this post I'll test a) H2O's function h2o.automl() that may help me automate the former and b) Thomas Lin Pedersen's library(lime) that may help clarify the latter. This post would never happen if not for the inspiration I got from two excellent blog posts: Shirin Glander's Lime presentation and Matt Dancho's HR data analysis. There's no hiding that this post is basically copy-catting their work, at least I'm standing on the shoulders of giants, hey! I'll use the powerful h2o.automl() function to optimise and choose the most accurate model classifying benign and malignant cancer cells from the Wisconsin dataset. First, let's load the data: Now, let's set up the local H2O instance… Finally, we can now use the famous h2o.automl() function and set the model up: set the target, feature names, training and validation set, as well as how long we want the algorithm to run for (for this you can use either max_runtime_secs argument, like I did here, or max_models, see h2o.automl() documentation for details.

Breast density classification is an essential part of breast cancer screening. Although a lot of prior work considered this problem as a task for learning algorithms, to our knowledge, all of them used small and not clinically realistic data both for training and evaluation of their models. In this work, we explore the limits of this task with a data set coming from over 200,000 breast cancer screening exams. We use this data to train and evaluate a strong convolutional neural network classifier. In a reader study, we find that our model can perform this task comparably to a human expert.