Chris Eliasmith, the director of the University of Waterloo's Centre for Theoretical Neuroscience, Canada, is trying to build a brain. Eliasmith's artificial model, Spaun, currently has just 2.5 million neurons to the human brain's 100 billion. But unlike more computationally demanding simulations, which have run for only a few simulated seconds, it's actually capable of doing something with them. "There's been an attitude of scale for scale's sake," Eliasmith explains. "But for us, the focus was discovering how neurones can be organised to produce behaviours, such as solving simple intelligence tasks."

Facebook could one day build facial gesture controls for its app thanks to the acquisition of a Carnegie Mellon University spinoff company called FacioMetrics. The startup made an app called Intraface that could detect seven different emotions in people's faces, but it's been removed from the app stores. The acquisition aligns with a surprising nugget of information Facebook slipped into a 32-bullet point briefing sent to TechCrunch this month. "Future applications of deep learning platform on mobile: Gesture-based controls, recognize facial expressions and perform related actions" It's not hard to imagine Facebook one day employing FacioMetrics' tech and its own AI to let you add a Like or one of its Wow/Haha/Angry/Sad emoji reactions by showing that emotion with your face. "How people share and communicate is changing and things like masks and other effects allow people to express themselves in fun and creative ways. We're excited to welcome the Faciometrics team who will help bring more fun effects to photos and videos and build even more engaging sharing experiences on Facebook."

In business you have to look at digital usage measurements as chess pieces--what pieces are on the chessboard, how they move, and what movements they cause downstream, can impact future decisions. That downstream view usually includes A/B and multivariate testing. Optimization is typically deployed to improve website elements that impact user experience. Changes in the optimization test platforms for Adobe and Google Analytics reflect the effect of machine learning on increasing the accuracy of test results and helping wmake better decisions. Adobe announced Auto-Target, a machine learning protocol that automates personalization-focused testing of elements to determine the preferred individual experience with media.

Raise your hand if you're a bit too optimistic when it comes to planning your day with a to-do list. If you're like me, chances are you write down too many tasks and always end up with unfinished tasks at the end of the day. Popular task-management service Todoist wants to help you reschedule your tasks and even out the work load using machine learning. Smart scheduling is a work in progress and will get better over time, but it sounds like a promising feature for intense Todoist users. "We created a couple of neural networks that can help our users with scheduling and rescheduling stuff. We think it's a pretty cool usage of AI," Todoist founder and CEO Amir Salihefendic told me.

There are approximately 1,100 known antimicrobial peptides (AMP) with diverse sequences that can permeate microbial membranes. To help discover the "blueprint" for natural AMP sequences, researchers from the University of Illinois at Urbana-Champaign and the University of California, Los Angeles, have developed a new machine learning approach to discover and design alpha-helical membrane active peptides based on their physicochemical properties. "In this work, we have trained a machine learning classifier--known as a support vector machine--to recognize membrane activity and experimentally calibrated the recognition metric by peptide synthesis and characterization," explained Andrew Ferguson, an assistant professor of materials science and engineering at Illinois. "We use machine learning to not only discover new membrane active peptides, but to also identify membrane activity in known peptides with previously defined functions leading us to discover membrane activity in diverse and unexpected peptide families. "Since getting cargo into a cell is important for many applications, we anticipate that this tool can have broad biomedical implications including in immunotherapy and in broad-spectrum membrane-active antimicrobial peptides to combat the rising incidence of drug resistance, design of cationic cell-penetrating peptides for nucleic acid transfection into cells, and in targeting and permeating anticancer therapeutics into tumors," added Ferguson, who was the senior computational investigator for the project. In this collaborative work, the Illinois researchers developed the computational innovations, with the experimental testing of the predictions accomplished at UCLA. The results, which highlight the difference between the efficacy of an antimicrobial and its recognizability as such, are surprising. "AMPs do not share a common core structure, but tend to be short, cationic, and amphiphilic," Ferguson said. "By training our machine learning classifier over a training set comprising peptides with known antimicrobial activity (hits) and decoy peptides with no activity (misses), the classifier learned the physical and chemical properties of a peptide that make for good membrane activity.

Why Is Crude Oil Weaker Early on November 16? Stay up-to-date on the topics you care about. We'll send you an email alert whenever a news article matches your alert term. It's free, and you can add new alerts at any time.

There's an arms race among public cloud providers to provide businesses with the best machine learning capabilities. Enterprises are increasingly interested in creating intelligent applications, and companies like Amazon, Microsoft and Google are rushing to help meet their needs. Google fired its latest salvo on Tuesday, announcing a set of enhancements to its existing suite of cloud machine-learning capabilities. The first was a new Jobs API aimed at helping match job applicants with the right openings. In addition, the company is slashing the prices on its Cloud Vision API and launching an enhanced version of its translation API.

That's an interesting question, and I try to answer this is a very general way. The tl;dr version of this is: Deep learning is essentially a set of techniques that help we to parameterize deep neural network structures, neural networks with many, many layers and parameters. And if we are interested, a more concrete example: Let's start with multi-layer perceptrons (MLPs)... On a tangent: The term "perceptron" in MLPs may be a bit confusing since we don't really want only linear neurons in our network. Using MLPs, we want to learn complex functions to solve non-linear problems. Thus, our network is conventionally composed of one or multiple "hidden" layers that connect the input and output layer.

I remember a time, before the Web, when you would look for relevant academic papers by reading large books with tiny fonts that would list all relevant work in a given area published in a given year. Of course, you could have just gone to the shelves and checked the research articles themselves but, by for a slow human being, this would have been just too time consuming. These large volumes contained nothing by "metadata": lists of article titles, authors, keywords… They were tremendously valuable to the researchers. One of the earliest applications of computers was to help manage document collections. In this sense, the Web and Google are very naturally applications for computers.

The next big thing in AI may not be getting a machine to perform a task--it might be requiring the machine to communicate why it took that action. For instance, if a robot decides to take a certain route across a warehouse, or a driverless car turns left instead of right, how do we know why it made that decision? According to Manuela Veloso, professor of computer science at Carnegie Mellon University, explainable AI is essential to building trust in our systems. Veloso, who works with co-bots (collaborative robots), programs the machines to verbalize their decision process. "We need to be able to question why programs are doing what they do," Veloso said.