Just before 9 am last Thursday, an unusual speed dating scene sprang up in San Francisco. A casually dressed crowd, mostly male, milled around a gilt-edged Beaux Arts ballroom on Nob Hill. Pairs and trios formed quickly, but not in search of romance. Ice breakers were direct: What's your favorite programming language? Which data analysis framework are you most expert in?

A prized attribute among law enforcement specialists, the expert ability to visually identify human faces can inform forensic investigations and help maintain safe border crossings, airports, and public spaces around the world. The field of forensic facial recognition depends on highly refined traits such as visual acuity, cognitive discrimination, memory recall, and elimination of bias. Humans, as well as computers running machine learning (ML) algorithms, possess these abilities. And it is the combination of the two--a human facial recognition expert teamed with a computer running ML analyses of facial image data--that provides the most accurate facial identification, according to a recent 2018 study in which Rama Chellappa, Distinguished University Professor and Minta Martin Professor of Engineering, and his team collaborated with researchers at the National Institute of Standards and Technology and the University of Texas at Dallas. Chellappa, who holds appointments in UMD's Departments of Electrical and Computer Engineering and Computer Science and Institute for Advanced Computer Studies, is not surprised by the study results.

Clinical Informatics tells us that: "Every year in the U.S., approximately 2 million patients participate in roughly 3000 clinical trials; six million patients are needed to meet U.S. recruitment goals. Consequently, up to 90% of trials are delayed or over budget". Experts blame the lack of data available - to both patients and researchers - to explain why only 5% of cancer patients, for example, end up enrolling in clinical trials. A study from Carnegie Mellon University and Albert Ludwig University in Germany predicts that "AI could cut the cost of drug discovery by about 70%" and Krishna Yeshwant, general partner at Google Ventures, estimates "AI would cut (clinical trial) costs by 90 percent." Artificial intelligence seems like the perfect solution, but Zikria Syed writes in MedCityNews that "clinical trial technologies haven't changed much since the current categories -- clinical trial management systems, electronic data capture, and interactive voice response, -- were established in the late 1990s." A recent Deloitte study also that tells us "a number of clinical trial activities still use the same processes as in the 1990s." In a sector that is usually at the forefront of technology – biotechnology - it is hard to believe this is happening. I spoke to six innovators who were tackling the massive problem head on – scientists and entrepreneurs working to bring clinical trials to the people who need them – to find out what they are doing to solve the serious innovation problem. The list of people is impressive for the diversity of solutions they're offering to clinical trials: Anna Huyghues-Despointes, Head of Strategy, Owkin; Simon Smith, Chief Growth Officer, BenchSci; Leila Pirhaji, Founder & CEO, ReviveMed; Shai Shen-Orr, Founder, Cytoreason; and Daniel Jamieson, CEO Biorelate and Gunjan Bhardwaj, Founder & CEO, Innoplexus. Additionally we spoke to consultant Dr. Chrysanthi Ainali, Co-Founder Dignosis and Instructor for the KNect365 Learning Course AI & Real World Evidence for Clinical Trials to ask her thoughts on the specific challenges AI startups in clinical trials face. "Healthcare brings great challenges for a technology company. It is inherently conservative and risk averse - Hippocratic Oath: 'first, do no harm'" says Simon Smith, Chief Growth Officer at Benchsci.

To demonstrate how easy it is to track people without their knowledge, we collected public images of people who worked near Bryant Park (available on their employers' websites, for the most part) and ran one day of footage through Amazon's commercial facial recognition service. Our system detected 2,750 faces from a nine-hour period (not necessarily unique people, since a person could be captured in multiple frames). It returned several possible identifications, including one frame matched to a head shot of Richard Madonna, a professor at the SUNY College of Optometry, with an 89 percent similarity score.

In seconds she gets a diagnosis of the disease affecting her plant and how best to manage it to boost her production. The farmer used an app on her phone based on TensorFlow, Google's Artificial Intelligence (AI) machine that the company opensourced to help developers create solutions to real-world problems. When people think of Artificial Intelligence, they most likely think of scenes from science fiction movies, but in reality, it applies to everyday life from virtual assistants to language translation on Google, says John Quinn, an AI researcher. Google now wants to position itself as an "AI first" company and with research centers across the globe in places such as Tokyo, Zurich, New York, and Paris. And last week, the technology company opened its first center in Africa in Ghana's capital city, Accra.

If you've heard of quantum computing, you might be excited about the possibility of applying it to machine learning applications. I work at Springboard, and we recently launched a machine learning bootcamp that includes a job guarantee. We want to make sure our graduates are exposed to cutting-edge machine learning applications -- so we put together this article as part of our research into the intersection of quantum computing and machine learning. Let's start by examining the difference between quantum computing and classical computing. In classical computing, your data is stored in physical bits and it is binary and mutually exhaustive: a bit is either in a 0 state or in a 1 state and it cannot be both at the same time.

To combat the robberies, Brazil's banks have invested in anti-theft technology, ranging from specialized ATMs to facial recognition cameras. When that fails or the costs become prohibitive, they have simply closed branches; as a result, some towns no longer have easy access to financial services in a country that already has a higher proportion of "unbanked" residents than either China or India.

Hundreds, maybe thousands, of dockworkers, residents and business owners are expected to pack the Board of Harbor Commissioners hearing Tuesday morning in an attempt to block a permit for infrastructure improvements at one Port of Los Angeles terminal that they fear would pave the way for the loss of high-paying jobs and the economic decline of surrounding communities. This small project -- about $1.5 million worth of electric charging stations, poles for Wi-Fi antenna and the like -- would be a precursor to a big change in San Pedro and harbor communities. APM Terminals and Danish shipping giant Maersk intend to automate operations over the next several years, potentially slashing the number of dockworkers in the future. These are middle-class jobs that can often pay workers without a college degree more than $100,000 a year. But, really, this fight goes far beyond APM Terminals.

A former electronics whiz kid, he has squandered his youth on dilettantish studies in physics and anthropology, followed by a series of botched get-rich-quick schemes. His parents are dead, his friends (if they exist) go unmentioned, and his employment consists of forex trading on an old laptop in his two-room apartment. He seems to leave home only to buy chocolate at a local newsstand or, once, after noticing a pain in his foot, to have an ingrown toenail removed, an apt literalization of his enervating self-involvement. Perhaps out of some desire for correction, Charlie sells his mother's house to finance the purchase of Adam, one of twenty-five cutting-edge androids built to serve as an "intellectual sparring partner, friend and factotum." The impulsive slacker is all too ready to exchange his birthright for a mess of wattage.

The deep image prior was recently introduced as a prior for natural images. It represents images as the output of a convolutional network with random inputs. For "inference", gradient descent is performed to adjust network parameters to make the output match observations. This approach yields good performance on a range of image reconstruction tasks. We show that the deep image prior is asymptotically equivalent to a stationary Gaussian process prior in the limit as the number of channels in each layer of the network goes to infinity, and derive the corresponding kernel. This informs a Bayesian approach to inference. We show that by conducting posterior inference using stochastic gradient Langevin we avoid the need for early stopping, which is a drawback of the current approach, and improve results for denoising and impainting tasks. We illustrate these intuitions on a number of 1D and 2D signal reconstruction tasks.