Visualization of high-dimensional data, such as text documents, is useful to map out the similarities among various data points. In the high-dimensional space, documents are commonly represented as bags of words, with dimensionality equal to the vocabulary size. Classical approaches to document visualization directly reduce this into visualizable two or three dimensions. Recent approaches consider an intermediate representation in topic space, between word space and visualization space, which preserves the semantics by topic modeling. While aiming for a good fit between the model parameters and the observed data, previous approaches have not considered the local consistency among data instances. We consider the problem of semantic visualization by jointly modeling topics and visualization on the intrinsic document manifold, modeled using a neighborhood graph. Each document has both a topic distribution and visualization coordinate. Specifically, we propose an unsupervised probabilistic model, called Semafore, which aims to preserve the manifold in the lower-dimensional spaces through a neighborhood regularization framework designed for the semantic visualization task. To validate the efficacy of Semafore, our comprehensive experiments on a number of real-life text datasets of news articles and Web pages show that the proposed methods outperform the state-of-the-art baselines on objective evaluation metrics.

Dynamic Bayesian networks (DBNs) are a general model for stochastic processes with partially observed states. Belief filtering in DBNs is the task of inferring the belief state (i.e. the probability distribution over process states) based on incomplete and noisy observations. This can be a hard problem in complex processes with large state spaces. In this article, we explore the idea of accelerating the filtering task by automatically exploiting causality in the process. We consider a specific type of causal relation, called passivity, which pertains to how state variables cause changes in other variables. We present the Passivity-based Selective Belief Filtering (PSBF) method, which maintains a factored belief representation and exploits passivity to perform selective updates over the belief factors. PSBF produces exact belief states under certain assumptions and approximate belief states otherwise, where the approximation error is bounded by the degree of uncertainty in the process. We show empirically, in synthetic processes with varying sizes and degrees of passivity, that PSBF is faster than several alternative methods while achieving competitive accuracy. Furthermore, we demonstrate how passivity occurs naturally in a complex system such as a multi-robot warehouse, and how PSBF can exploit this to accelerate the filtering task.

The design of multiple experiments is commonly undertaken via suboptimal strategies, such as batch (open-loop) design that omits feedback or greedy (myopic) design that does not account for future effects. This paper introduces new strategies for the optimal design of sequential experiments. First, we rigorously formulate the general sequential optimal experimental design (sOED) problem as a dynamic program. Batch and greedy designs are shown to result from special cases of this formulation. We then focus on sOED for parameter inference, adopting a Bayesian formulation with an information theoretic design objective. To make the problem tractable, we develop new numerical approaches for nonlinear design with continuous parameter, design, and observation spaces. We approximate the optimal policy by using backward induction with regression to construct and refine value function approximations in the dynamic program. The proposed algorithm iteratively generates trajectories via exploration and exploitation to improve approximation accuracy in frequently visited regions of the state space. Numerical results are verified against analytical solutions in a linear-Gaussian setting. Advantages over batch and greedy design are then demonstrated on a nonlinear source inversion problem where we seek an optimal policy for sequential sensing.

Vice President Joe Biden will push for international cooperation in the fight against cancer in a speech at the Vatican. The vice president's office says the address on Friday will look at global research partnerships and will describe how his cancer "moonshot" project may have an international impact. Biden is due to speak at an international conference on breakthroughs in regenerative medicine. The gathering of doctors, patients and researchers is hosted by the Pontifical Council for Culture and the Stem for Life Foundation. Biden's office says the vice president will visit with Pope Francis during the stop.

Stanford's five-foot "virtual diver" was originally built for studying coral reefs in the Red Sea where a delicate touch is necessary, but the depths go well beyond the range of meat-based divers. The "tail" section contains the merbot's onboard batteries, computers and array of eight thrusters, but it is the front half that looks distinctly humanoid with two eyes for stereoscopic vision and two nimble, articulated arms. Those arms are what make OceanOne ideal for fragile reef environments or priceless shipwrecks like La Lune, which sank off the coast of France over 350 years ago and hasn't been touched until now. Force sensors in each wrist transmit haptic feedback to the pilot, allowing them to feel the object's weight while staying high and dry on a dive ship. The robot's "brain" works with the tactile sensors to ensure the hands don't crush fragile objects, while the navigation system can automatically keep the body steady in turbulent seas.

It's that time of year: The most prepared of you are carboloading after weeks of training for Marathon Monday, while the least prepared are scrambling to finish your taxes. Either way, surely you'll need some breaks in the pasta and accounting. Should you take your kids to Disney's live-action/CGI remake of its beloved 1967 film "The Jungle Book"? Absolutely, says Ty Burr, who gives three stars to this movie placing "talking animals of almost tactile musculature and movement" in a lush jungle landscape; it holds up right until its overly frenetic final scenes. The jungle beasts are voiced by the likes of Ben Kingsley, Lupita Nyong'o, Bill Murray, Idris Elba, and Scarlett Johanssen, and newcomer Neel Sethi is charming as Mowgli.

A brood of sidewalk drones could be rolling around the nation's capital within a year, if a District of Columbia Council member has her way. Executives from Starship Technologies, with roots in Estonia and London, say their goal is to unleash a platoon of "smart, friendly robots" that will ply sidewalks along with pedestrians to make local deliveries of groceries or small packages "almost free." The company is led by Skype co-founders Ahti Heinla and Janus Friis, and launched the effort in November. Councilwoman Mary Cheh and company officials sought to make a splash by promising one of the squat vehicles on Wednesday would deliver legislation to the council authorizing self-driving delivery robots. The little white device, which looks like an ice chest rolling on six wagon wheels, did indeed scoot its way into Council Secretary Nyasha Smith's office with the three-page bill in its compartment and reporters on its tail.

The findings of a recent survey from Nuance Communications into consumer preferences and expectations around customer self-service reveals that a significant number of consumers (89%) prefer, and even expect, to have conversational interaction. For 87%, a positive interaction with a company determines whether or not they will continue their relationship. With chat bots to the front of mind following Facebook's recent announcement, it is interesting to note from Nuance's survey that 89% of consumers "want to engage with virtual assistants to quickly find information instead of searching through Web pages or a mobile phone app on their own". The same need is reflected in consumers' desire to engage with automated phone systems that allow them to speak naturally. Nuance recently announced that one of the leading financial institutes in Sweden, Estonia, Latvia and Lithuania is now using Nuance Nina, and intelligent virtual assistant that delivers "a human-like, conversational customer service experience" to assist customers to quickly and easily access information.

A new article in ScienceDaily reports, "A century ago, more than 60,000 tigers roamed the wild. Today, the worldwide estimate has dwindled to around 3,200. Poaching is one of the main drivers of this precipitous drop. Whether killed for skins, medicine or trophy hunting, humans have pushed tigers to near-extinction. The same applies to other large animal species like elephants and rhinoceros that play unique and crucial roles in the ecosystems where they live. Human patrols serve as the most direct form of protection of endangered animals, especially in large national parks. However, protection agencies have limited resources for patrols. With support from the National Science Foundation (NSF) and the Army Research Office, researchers are using artificial intelligence (AI) and game theory to solve poaching, illegal logging and other problems worldwide, in collaboration with researchers and conservationists in the U.S., Singapore, Netherlands and Malaysia."

The good news: The Loch Ness Monster has been captured on sonar by an underwater robot operated by the British division of Norway's Kongsberg Maritime. The bad news: "Nessie" is a prop from a Sherlock Holmes film that sank in the loch in 1969. The monstrous model was long thought lost until it was discovered this week by the Munin Autonomous Underwater Vehicle (AUV) as part of an underwater survey of the loch for The Loch Ness Project and VisitScotland. There have been sporadic sightings of what is purported to be the Loch Ness Monster since the first recorded encounter by St Columba in 565 AD. After a supposed photograph was taken in 1933, public interest in some sort of large, dinosaur-like creature making its home in the Highlands skyrocketed, and in the decades since the loch has been subjected to sonar scans, submersible hunts, hydrophone surveys, and enough photographs taken above and below the surface to wallpaper the Grand Canyon.