Data Science newsletter – July 31, 2018

GROUP CURATION: N/A

Physicians at the University of California, Irvine and UCI Health System have launched the UCI Center for Artificial Intelligence in Diagnostic Medicine, which seeks to advance patient care, improve health outcomes and lower costs by leveraging machine learning technology in all areas of healthcare.

Led by Peter D. Chang, MD, and Daniel S. Chow, MD, neuroradiologists in the Department of Radiological Sciences, UCI School of Medicine, the center is a cross-specialty initiative with a specific focus on developing and applying deep learning neural networks to healthcare applications, such as diagnostics, disease prediction and therapy planning.

The growing popularity of always-listening devices is driving a migration toward the voice user interface. From smart speakers and hearables to robots, smart home systems, and TV remotes, we enjoy using voice to communicate with the electronic devices in our lives because it feels natural. Many of us also feel more secure when we have another set of virtual ears tuned into our sleeping baby or to our home-security system. As electronics-component suppliers who are already profiting from insatiable demand for voice-enabled everything, MEMS microphone suppliers are closely attuned to the potential benefits and risks of always-listening devices. Where do they think we are going with always-listening devices and to what degree are we risking our private data by using them?

According to IHS Markit, we will have more than 5 billion consumer devices using digital assistants in 2018, growing to nearly 8 billion by 2021. What’s driving this growth?

Twitter is partnering with two groups of academic researchers to figure out how to measure the health of conversations happening on the platform. … The first team, led by scholars from Leiden University, will look at how echo chambers form and their effect, as well as the difference between incivility and intolerance within Twitter conversations. The team — including Dr. Rebekah Tromble, Assistant Professor of Political Science at Leiden University, Dr. Michael Meffert at Leiden, Dr. Patricia Rossini and Dr. Jennifer Stromer-Galley at Syracuse University, Dr. Nava Tintarev at Delft University of Technology, and Dr. Dirk Hovy at Bocconi University — has found in past research that echo chambers can cause hostility and promote resentment towards those not having the same conversation. … Professor Miles Hewstone and John Gallacher at The University of Oxford, in partnership with Marc Heerdink at the University of Amsterdam, will lead the second research project.

The University of New South Wales (UNSW) has developed a free mapping tool, powered by Google Earth and NASA, that shaves weeks from the time taken to monitor and assess changes to landscapes and ecosystems.

The Remap program uses machine learning to comb through NASA’s Landsat imagery, which has been updated fortnightly since the 1970s, combining it with climate data to track large scale changes in different ecosystems.

A new way of arranging advanced computer components called memristors on a chip could enable them to be used for general computing, which could cut energy consumption by a factor of 100.

This would improve performance in low power environments such as smartphones or make for more efficient supercomputers, says a University of Michigan researcher.

“Historically, the semiconductor industry has improved performance by making devices faster. But although the processors and memories are very fast, they can’t be efficient because they have to wait for data to come in and out,” said Wei Lu, U-M professor of electrical and computer engineering and co-founder of memristor startup Crossbar Inc.

While some Silicon Valley giants grapple with the ethics of offering cutting-edge artificial intelligence technology to military and law enforcement agencies with histories of abuse, Amazon, apparently, has no reservations.

When asked about the culture “gap” between Amazon employees — who have protested the sale of facial recognition technology to law enforcement — and the company’s “executive level” interests, Teresa Carlson, vice president of the worldwide public sector of Amazon Web Services, was frank. “We are committed to our customer, and we are unwaveringly committed to the U.S. government and the governments we work with around the world,” Carlson declared at the Aspen Security Forum on July 20 in Colorado.

Carlson’s remarks, largely unreported outside of a mention in a technology trade journal, mark the lengthiest public discussion in recent months of Amazon’s role as a technology provider for military and law enforcement, which has been a source of substantial controversy for the company.

Recent advances in automating and digitizing financial services has largely changed how we use technology to make fiscal decisions. As infrastructure and operations are changing rapidly — recent reports show $31 billion was invested in the sector in 2017 — reinventing global financial technology poses many challenges and potential rewards.

To address these issues, MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) recently launched a research-industry collaboration focused on creating financial technologies that will be able to open up new business models, gain new data insights, and improve security.

The initiative will span topics that include artificial intelligence, cryptocurrencies, blockchain foundation and applications, machine learning, multi-party computation for superior security and privacy, data management and analytics, natural language processing, and cyber-risk management, among others.

Bacteria go extinct at substantial rates, although appear to avoid the mass extinctions that have hit larger forms of life on Earth, according to new research from the University of British Columbia (UBC), Caltech, and Lawrence Berkeley National Laboratory. The finding contradicts widely held scientific thinking that microbe taxa, because of their very large populations, rarely die off.

The study, published today in Nature Ecology and Evolution, used massive DNA sequencing and big data analysis to create the first evolutionary tree encompassing a large fraction of Earth’s bacteria over the past billion years.

“Bacteria rarely fossilize, so we know very little about how the microbial landscape has evolved over time,” says Stilianos Louca, a researcher with UBC’s Biodiversity Research Centre who led the study. “Sequencing and math helped us fill in the bacterial family tree, map how they’ve diversified over time, and uncover their extinctions.”

Days after the ACLU released a damning report on Amazon’s face recognition product Rekognition, Amazon’s general manager of AI, Dr. Matt Wood, countered its findings in a blog post. The ACLU used Rekognition to scan the faces of all 535 members of Congress, finding the software mistook 28 of them for suspected criminals. Dr. Wood notes first that the ACLU doesn’t reveal its methodology or dataset in the report, then punctuates Amazon’s original response—that it encourages higher confidence thresholds for law enforcement.

Conspicuously missing from the blog, however, was a specific rebuttal to the enormous racial disparity uncovered by the ACLU. For Congress as a whole, the error rate was only five percent, but for non-white members of Congress, the error rate was 39 percent. When one centers this racial disparity, as the blog post does not, Dr. Wood’s final point reads as an argument against face recognition use by law enforcement wholesale.

The 2017 economic census leadership team endorsed a number of innovative updates, each introducing a new set of statistical and production challenges. For example, in 2017, data collection will be primarily via the web, instead of mailing out paper questionnaires (as done in all previous censuses). Standard unit response rates will be released for the first time with the 2017 economic census, as will imputation rates for key statistics. Variability estimates for selected sample-based statistics will be published for the first time, as well. These last two sets of measures provide unprecedented transparency on the data quality for this large and widely used program.

And beginning in 2017, the economic census will use the North American Product Classification System (NAPCS) to produce economy-wide product tabulations. NAPCS is a comprehensive, market- or demand-based, hierarchical classification system for products (goods and services) developed and used by Canada, Mexico, and the United States (see http://www.census.gov/eos/www/napcs/more.html). Using this new classification system helps standardize vital economic statistics that measure the growth of output, prices, productivity, and trade.

Big news today from NIH announcing their new initiative to develop and test new ways to best implement cloud services in support of biomedical research. Called STRIDES for “Science and Technology Research Infrastructure for Discovery, Experimentation and Sustainability,” the initiative will allow NIH to explore the use of cloud environments to streamline NIH data use. By partnering with commercial cloud service providers, NIH expects to improve access to biomedical data and provide cost-effective cloud infrastructure, data storage, computation, and machine learning services for NIH and NIH-supported investigators.

I thank the many NLM staff members who contributed key knowledge to help shape this initiative, and I’m confident that NLM will have many opportunities to impact STRIDES’ success, whether by devising discovery mechanisms, developing tools for investigators, assigning metadata to incoming data sets, or developing effective ways to link those data sets to related publications.

Uber is shuttering its self-driving trucks unit, a beleaguered program borne out of the company’s controversial multi-million acquisition of Otto nearly two years ago.

The company said Monday that Uber Advanced Technologies Group will stop development of self-driving trucks and instead focus its efforts on self-driving cars.

“We recently took the important step of returning to public roads in Pittsburgh, and as we look to continue that momentum, we believe having our entire team’s energy and expertise focused on this effort is the best path forward,” Eric Meyhofer, head of Uber Advanced Technologies Group, said in an emailed statement.

China’s Ministry of Foreign Affairs has turned to artificial intelligence (AI) and “big data” to help the country make diplomatic decisions.

“Cutting-edge technology, including big data and artificial intelligence, is causing profound changes to the way people work and live. The applications in many industries and sectors are increasing on a daily basis,” a Chinese foreign ministry spokesman said, the South China Morning Post reported on July 30. “[We] will actively adapt to the trend and explore the use of emerging technology for work enhancement and improvement.”

Today, the U.S. Department of Energy (DOE) announced $10 million in funding for 13 projects aimed at further enhancing one of the world’s most sophisticated computer models for understanding weather and climate patterns.

The DOE-supported Energy Exascale Earth System Model (E3SM) seeks to provide more accurate and higher-resolution representation of weather and climate events by taking advantage of the cutting-edge supercomputing facilities at DOE National Laboratories.

“Modeling the Earth system in all its complexity represents an enormous challenge,” said U.S. Secretary of Energy Rick Perry. “The supercomputers at DOE National Laboratories are playing a critical role in this effort—a role that will grow in importance as we move into the era of next-generation exascale computing.”

Waymo has been working with a group of early riders in its test program in the Phoenix metro area for over a year and it is learning how they use their autonomous ride-hailing service.

Building on what they learned, Waymo is now announcing new partnerships to help riders better make use of their autonomous service, including a partnership with Walmart.

Waymo’s fleet of test vehicles across the U.S. is now driving more than 24,000 miles daily, many of them with actual customers in Phoenix.

Arlington, VA August 29. “This year at the CERT Data Science in Cybersecurity Symposium, we focus on the topic of metadata.” [free, registration required]

New York, NY September 23. “Held annually during the United Nations General Assembly week, the Summit unites a lively community of global citizens and progressive thought leaders around a common theme: #2030NOW.” [$$$]

Atlanta, GA September 12, starting at 12:15 p.m. in Marcus Nanotechnology Building, Georgia Tech. [free]

Vancouver, BC, Canada October 2-7. “Tentative list of all long papers accepted for RecSys 2018 (in alphabetical order). Please note that this list may be subject to change.” [$$$]

Atlanta, GA September 13. “The AWS Summit is a free event designed to bring together the cloud computing community to connect, collaborate, and learn about AWS. Learn by attending sessions ranging in technical depth from introduction to advanced.” [free, registration required]

Integrated information theory provides a mathematical framework to fully characterize the cause-effect structure of a physical system. Here, we introduce PyPhi, a Python software package that implements this framework for causal analysis and unfolds the full cause-effect structure of discrete dynamical systems of binary elements. The software allows users to easily study these structures, serves as an up-to-date reference implementation of the formalisms of integrated information theory, and has been applied in research on complexity, emergence, and certain biological questions. We first provide an overview of the main algorithm and demonstrate PyPhi’s functionality in the course of analyzing an example system, and then describe details of the algorithm’s design and implementation. PyPhi can be installed with Python’s package manager via the command ‘pip install pyphi’ on Linux and macOS systems equipped with Python 3.4 or higher. PyPhi is open-source and licensed under the GPLv3; the source code is hosted on GitHub at https://github.com/wmayner/pyphi. Comprehensive and continually-updated documentation is available at https://pyphi.readthedocs.io. The pyphi-users mailing list can be joined at https://groups.google.com/forum/#!forum/pyphi-users. A web-based graphical interface to the software is available at http://integratedinformationtheory.org/calculate.html.

“Congratulations, you’ve published your paper! As focus on science communication (scicomm) continues to rise in the academic community, social media is a natural fit for promoting work to the media, public, or policy makers. It also provides a space for open, two-way dialogue. The below list of tips will give you a great jumpstart in learning to utilize social media to engage the public in your research.”

“1. Using social media to promote your work can start well before your paper publishes.”