The sizes of Markov equivalence classes of directed acyclic graphs play important roles in measuring the uncertainty and complexity in causal learning. A Markov equivalence class can be represented by an essential graph and its undirected subgraphs determine the size of the class. In this paper, we develop a method to derive the formulas for counting the sizes of Markov equivalence classes. We first introduce a new concept of core graph. The size of a Markov equivalence class of interest is a polynomial of the number of vertices given its core graph. Then, we discuss the recursive and explicit formula of the polynomial, and provide an algorithm to derive the size formula via symbolic computation for any given core graph. The proposed size formula derivation sheds light on the relationships between the size of a Markov equivalence class and its representation graph, and makes size counting efficient, even when the essential graphs contain non-sparse undirected subgraphs.

The question of how to parallelize the stochastic gradient descent (SGD) method has received much attention in the literature. In this paper, we focus instead on batch methods that use a sizeable fraction of the training set at each iteration to facilitate parallelism, and that employ second-order information. In order to improve the learning process, we follow a multi-batch approach in which the batch changes at each iteration. This can cause difficulties because L-BFGS employs gradient differences to update the Hessian approximations, and when these gradients are computed using different data points the process can be unstable. This paper shows how to perform stable quasi-Newton updating in the multi-batch setting, illustrates the behavior of the algorithm in a distributed computing platform, and studies its convergence properties for both the convex and nonconvex cases.

The highly variable dynamics of neocortical circuits observed in vivo have been hypothesized to represent a signature of ongoing stochastic inference but stand in apparent contrast to the deterministic response of neurons measured in vitro. Based on a propagation of the membrane autocorrelation across spike bursts, we provide an analytical derivation of the neural activation function that holds for a large parameter space, including the high-conductance state. On this basis, we show how an ensemble of leaky integrate-and-fire neurons with conductance-based synapses embedded in a spiking environment can attain the correct firing statistics for sampling from a well-defined target distribution. For recurrent networks, we examine convergence toward stationarity in computer simulations and demonstrate sample-based Bayesian inference in a mixed graphical model. This points to a new computational role of high-conductance states and establishes a rigorous link between deterministic neuron models and functional stochastic dynamics on the network level.

The company's total income has increased from Rs 13,198.6 crore for the quarter ended September 30, 2015 to Rs 14,407.3 The company has posted a net profit after taxes, minority interest and share of profit of associates of Rs 2070 crore for the quarter ended September 30, 2016 as compared to Rs 2241 crore for the quarter ended September 30, 2015. Total Income has increased from Rs 13198.6 crore for the quarter ended September 30, 2015 to Rs 14407.3 On a standalone bais, the company has posted a net profit of Rs 1932 crore for the quarter ended September 30, 2016 as compared to Rs 2153 crore for the quarter ended September 30, 2015. Total Income has increased from Rs 11725 crore for the quarter ended September 30, 2015 to Rs. 12101 crore for the quarter ended September 30, 2016.

Mphasis, a leading IT services and solutions provider, today announced the launch of DigiOps driven by InfraGenie, an intelligent automation platform (IAP), powered by Arago, a pioneer in artificial intelligence (AI) and leader in intelligent IT automation. DigiOps delivers solutions by reducing manual effort across IT functions. InfraGenie intelligently predicts incidents before they arise so that companies have a reliable and consistent way of solving errors in their industry-specific IT operations. This smart infrastructure solution unites the proficiencies of advanced analytics ("prescriptive") and artificial intelligence based automation to offer resolutions for all types of infrastructure related events. Through InfraGenie, Mphasis will bring both automation and analytics together to reliably and consistently identify, predict and resolve the infrastructure problems of today's complex hybrid IT environment.

To cite this article: Gina Neff. The biggest challenge for the use of "big data" in health care is social, not technical. Data-intensive approaches to medicine based on predictive modeling hold enormous potential for solving some of the biggest and most intractable problems of health care. The challenge now is figuring out how people, both patients and providers, will actually use data in practice. "I FOUND THE BUZZ AS FEVERISHLY LOUD AROUND HEALTH INFORMATION INNOVATION AS IT WAS DURING MY RESEARCH ON THE FIRST DOT-COM BOOM." To understand how data-intensive solutions could have an impact on health care, our research team talked to frontline providers in impoverished and rural areas, technology enthusiasts in mobile health and health IT startups, clinicians and researchers in major research hospitals, Quantified Self members at data-driven meetup presentations of massive amounts of tracking data, and attendees at the growing number of conferences for health technology and innovation up and down both coasts. I found the buzz as feverishly loud around health information innovation as it was during my research on the first dot-com boom. One of our findings from this research seems at first blush so obvious that it is hard to believe it has been overlooked in the design and implementation of health-care innovation technologies.

I had the honor to be one of the warmup acts for President Obama at the White House Frontiers Conference at Carnegie Mellon University in Pittsburgh. Here is the prepared text and slides from the talk I delivered there. As you'll see if you watch the video, what I ended up saying isn't exactly what I had written out in advance, but it is reasonably close. Hearing that Bob Dylan just won the Nobel Prize for Literature, how could I not begin this talk with his famous line, "Something is happening here, but you don't know what it is, do you, Mr. Jones?" The future is full of amazing things.

Autonomous robots can perform actions or complete tasks with a high degree of autonomy, which makes them ideal for applications such as space exploration or cleaning your living room carpet. Mobile robots are capable of moving from place to place. Researchers say a wearable tattoo can deliver accurate blood-alcohol level readouts to its wearer's smartphone. Put these capabilities together and you got a powerful machine that can handle lots of tasks in industrial environments such as factories, as well as in hospitals, hotels, and other areas. And, in fact, one of the more prominent trends in robotics today is the growing popularity of autonomous mobile robots (AMRs), with new vendors jumping into the market and sales on the rise.

At the end of Star Wars Episode V: The Empire Strikes Back, Luke Skywalker feels when a needle pricks his newly-installed bionic hand. Researchers report today in the journal Science Translational Medicine that they can do something similar: stimulating regions of a human test subject's brain with electrodes can recreate the perception of touch in a robotic hand. This year, about 280,000 people in the United States alone are living with a spinal cord injury, according to the National Spinal Cord Injury Statistical Center. Depending on the severity, damaged nerve connections lead to effects ranging from a partial loss of feeling to complete loss of motion in different limbs. "If you lose that sense of touch, you have a really difficult time" grabbing, holding, and manipulating different objects, says Richard Gaunt, a neuroengineer at the University of Pittsburgh who works on touch feedback for prosthetics.

Microsoft's artificial intelligence (AI) technology can now recognize conversational speech slightly better than humans who do so professionally, according to recently-released research from the company. Microsoft recently got its AI's error rate in understanding speech down to about 5.9% from 6.3%, which puts it slightly below the human error rate, which is also close to 5.9% "We [improved] on our recently reported conversational speech recognition system by about 0.4%, and now exceed human performance by a small margin," the report stated. This news comes a mere month after Microsoft announced that it had reached an error rate of 6.3%, at the time setting a record among its peers. However, Microsoft's research also noted that error rates of human transcribers can vary between 4.1% to 9.6%, depending on how carefully they perform the transcription. SEE: AI experts weigh in on Microsoft CEO's 10 new rules for artificial intelligence Still, the closeness in quality to human transcription is impressive.