Yihong Wu creates new tools for digging into data

By Jim Shelton

Somewhere, buried deep inside mountains of information, awaits the human dimension of data. It’s the small subset of material that, when properly selected, sheds light on something important, such as public policy or DNA sequencing.

This is the scientific territory where Yihong Wu, a Yale assistant professor of statistics and data science, has set up shop. He’s made it his mission to find communities and networks within high-dimensional data.

All scientific fields deal with data, and more of it pours in all the time. But it’s not easy to make sense of it unless you do it in a principled, grounded way,” Wu says.

Wu is part of a wave of new faculty joining the Department of Statistics and Data Science, as Yale continues to weave data science into the fabric of campus research in all disciplines. The University Science Strategy Committee recently named data science as a top priority and recommended that Yale invest in a university-wide initiative to integrate data science and mathematical modeling research across campus.

Wu joined the Yale faculty in 2016. Earlier this year, he earned a prestigious Sloan Research Fellowship, an award aimed at helping promising, early-career scientists.

Yihong’s work lies at the intersection of high-dimensional statistics, information theory, and computer science,” said Harrison Zhou, professor and chair of the Department of Statistics and Data Science. “He has made fundamental contributions in the important problem of estimating the number of unseen symbols in a population.”

Researchers have been working on this for generations, Zhou noted. In 1943, Ronald Fisher, Alexander Steven Corbet, and Carrington Williams wrote a seminal study about estimating species diversity, based on moth and butterfly collections; in 1976, Bradley Efron and Ronald Thisted came up with an estimate of William Shakespeare’s vocabulary based on a dataset of his recorded works.

In today’s research, in virtually every discipline, there is the variable of volume: a flood of data that streams in continuously. This material offers a wealth of possibilities, but it also poses problems. How do you store the information? How do you sift through it in ways that are financially feasible and can be done in a reasonable amount of time?

Wu tries to answer these questions with two guiding ideas in mind: design algorithms that give you provable, guaranteed results and get the information quickly.

The starting point is always a good statistical model on which we can determine the optimal procedure,” Wu said. “However, due to the combinatorial nature of the problem, it might be computationally expensive to solve.”

The problems run the gamut of academic, business, and social inquiry. For instance, Wu might be looking at data about protein interactions in the human body in order to select which proteins can help create new medicines. Or he might be analyzing ways to improve online social networks by incorporating certain user data into the decision-making process.

One approach to such challenges, he said, is to “relax” the problem. Wu and his collaborators sometimes use optimization techniques called convex relaxations to solve a relaxed version of the original problem. In other situations, Wu uses “belief propagation,” an iterative algorithm normally used in statistical physics that essentially passes messages back and forth to fill in gaps in information.

In my research, I focus on both the theoretical and the algorithmic aspects of statistical problems,” Wu said. “For me, good research is achieved by finding methodologies that are theoretically grounded and computationally efficient. Equally important to me is proving the statistical optimality of the methods I propose.”

A good example of this is Wu’s work that revisited the classic problem of predicting the number of unseen species based on a collection of samples. Published in the Proceedings of the National Academy of Sciences, the research broke new ground in making such estimates.

Previous studies had shown how to estimate the number of species for a population more than twice the size of the sample — but without a provable guarantee of accuracy. Wu and collaborators Alon Orlitsky and Ananda Theertha Suresh not only provided that provable guarantee, but they also expanded the size of the population that could be examined.

The possibilities for Wu’s research are equally expansive. Finding communities within data will have implications for looking at voting records, DNA chains, the blogosphere, traffic patterns, and an array of other data. Wu said the task requires expertise in both theory and practice, which is why he came to Yale.

The main attraction to me was the people we have here,” he said. “There is a very common interest in foundational studies, but with a firm focus on practical applications. You want your research to be useful, to have impact.”

Source: Yale News


Public Service Baked In

Public service seems to be baked into the DNA of the Marconi Fellows and Young Scholars.

I thought about this at a meeting we held with the FCC in Washington DC in March. Our board was convening in DC and we offered to meet with FCC staff on any topic they wanted to discuss. We did this previously in 2014, in a mutually beneficial encounter.

The conversation this year, which included John Cioffi, Guilhem de Valicourt, Zvi Galil, Joseph Kakande, Bob Tkach, and me, focused on how the agency could more effectively monitor communications reliability and performance and how the agency might improve the public emergency alerting system. For several hours, we exchanged ideas with FCC staff members (not commissioners) in an informal meeting free of any political overtones. These issues have real ramifications for the public, as recent natural disasters have demonstrated, so improving the FCC’s ability to do its job could have important public benefits.

The FCC staff members agree. They recently requested a second meeting, this time to discuss applying blockchain technology to the communications sector–primarily for supply chain risk mitigation, but also to improve other long-known problems like secure inter-domain routing, secure naming, and inventory management. Several of our Fellows and Young Scholars have agreed to speak with them.  I am still skeptical of the hyperbole surrounding blockchain technology, so this is an important conversation.

This is but one example of how our board, Fellows, and Young Scholars lend their time, energy and intellect to efforts that benefit the public. That’s why we plan to include occasional features on their activities. Not all of them are technology-related. For example, Len Kleinrock has supported the Community Karate Foundation (he’s a black belt) for decades, promoting the health benefits of karate training.

Bob Lucky, whose history of advisory and technical committee memberships takes the better part of a page to list, has now turned to a more local challenge—redeveloping an abandoned US Army base, Fort Monmouth, near his home in New Jersey. He spent four years chairing the planning effort to turn the base into a new town, helped obtain federal government approval, and now chairs the redevelopment.

Like many of you, Marty Cooper has devoted substantial time to serving on advisory committees and boards at universities and government agencies. He is a member of the FCC Technological Advisory Council, chairing their antenna working group, a Trustee on the board of the Illinois Institute of Technology, Advisor to the Dean of Engineering at UC San Diego, and a charter member of TIE International, a group which mentors and honors young entrepreneurs. Yet he still finds time to serve on the board of the Cinequest Film Festival.

Our Young Scholars continue this tradition serving in both technical and social impact capacities.  Eric Plum supports student research projects at his former secondary school in Germany, supervising projects and building and maintaining a bilingual project database that has helped 166 student teams win 379 science competitions at the regional, state, national and global levels.

Junwen Zhang is on the technical subcommittee of the Optical Fiber Communication Conference (OFC), sponsored by the Optical Society (OSA) and on the technical subcommittee of the Photonics West Conference sponsored by SPIE, the international society for optics and photonics.

Salman Baset represents IBM in external efforts to coalesce on an industry point of view on identity so that blockchain technology can support the 1.1B people in the world who have no officially recognized identity, helping to meet the UN’s 2030 Sustainable Development goal of having a legal identity for each person on the planet.

I have written before about our Young Scholars’ Celestini Program, built to grow technical capability among students in developing countries by helping them apply technology to incubate solutions to problems that impact their local communities. Aakanksha Chowdhery, Guilhem de Valicourt and Joseph Kakande have spent countless hours starting this program in India and Uganda and it will expand this year to Colombia, Ghana and Rwanda.

This is only a tiny sampling of the interesting and important ways the Marconi Society continues to “benefit humanity.” I look forward to hearing about the service of others and we will feature more examples in the future. In the meantime, thanks for all that you do to keep the spirit of Guglielmo Marconi alive.

Empowering Students to Change Local Lives With Technology: What We Have Learned So Far

By Dr. Dorothy Okello and Paula Reinman 

“While technology is important, it’s what we do with it that truly matters.”
Muhammad Yunis – Founder, Grameen Bank

Two years ago, the Marconi Society Young Scholars, led by Dr. Joseph Kakande, created the Celestini Program to encourage undergraduate engineering students in developing countries to use technology to solve problems that are important in their local areas. The Program started in Uganda, expanded to India in 2017 and will launch in China, Colombia, Ghana and Rwanda in 2018.

With mentoring, support and global connections provided by the Young Scholars, the Marconi Society and other supporters including Google, IEEE, Nokia-Bell Labs, the Samueli Foundation and Vint and Sigrid Cerf, the Uganda project is now entering its third year. To date, 22 students have engaged in the project, as well as local OB GYNs and midwives.

Based on the positive feedback and results we have seen in Uganda and India, the Marconi Society will implement solutions and expand the program to benefit more people. Here are some of the most important things we have learned so far.

Makerere University and Maternal Mortality

The Uganda Celestini Project is hosted at Makerere University through netLabs! Uganda (netLabs!UG), a research Centre of Excellence in communications and networking technologies that is led by Dr. Dorothy Okello. NetLabs!UG is hosted by the Department of Electrical and Computer Engineering under the College of Engineering, Design, Art and Technology (CEDAT). NetLabs!UG’s goal of a balanced mix of basic research, applied research and commercialization is a perfect fit for the Celestini Program’s strategy of incubating technology to solve critical local issues.

When Dr. Kakande approached her about the idea of starting a Celestini Project at Makerere, Dr. Okello was immediately enthused. “I loved the idea,” she says. “I want to expose students to international partnerships and collaboration. I was excited about applying telecommunications to a health care problem in a developing country. It is not enough to be good at programming – our students need to think about how their technical prowess can help othersand healthcare is a great application area for that.”

The netLabs!UG team selected a potentially life-changing issue: identifying and treating pregnant women with hypertension.

Uganda has the 37thhighest maternal mortality rate in the world. Hypertension, a condition that is easy to identify and treat, is the second most common cause of maternal mortality in sub-Saharan Africa.

While the cure is known, the mortality rate is caused by lack of awareness among pregnant women and the fact that many of these women live in rural areas far away from health centers.

Dr. Okello has run this program twice with students and is now gearing up for a third time, applying what she has learned in each new session. “One of my main expectations was for students to be able to work on a real problem that is important to Ugandans. We have been able to do that – twice,” she says.



How it Works

After selecting maternal mortality due to hypertension as the problem to solve, the students in the first year program moved immediately into developing applications and tools to address the issue. They quickly created and iterated on a mobile application to track the health of pregnant women and identify the potential for hypertensive disorder, as well as a database to track patients, generate reports for doctors and store patient data.

In the second year, Dr. Okello focused on integrating user input into the entire process, starting with the solution design. The second year session kicked off with a design-thinking workshop with 33 participants from local women’s health groups and from the University, including students beyond those in the Celestini Project.

“I opened the design-thinking workshop to additional students because I believe that all students can benefit from this problem solving approach,” says Dr. Okello. “The workshop provides the front-line practitioner perspectives that help students test and refine their assumptions and it gives students contacts within the stakeholder community that they are serving.”

For example, students decided to use a smart bracelet to allow health care professionals to remotely monitor pregnant women for hypertension. The design-thinking workshop surfaced the high costs of the smart bracelets, along with the extra work for medical staff created by the reports from each smart bracelet. Based on this, the team decided to provide bracelets only to high-risk patients, containing the cost and generating a level of data that the health care professionals could review in a timely and high quality way.

Students are also developing a nationwide telemedicine assessment to quickly analyze a given clinic or hospital setting to determine whether it has capabilities to support telemedicine. From the design-thinking workshop, students learned that telemedicine is an extremely broad term and that they need to understand the type of facilities available based on the five-level health center model (with options ranging from village health teams to urban hospitals) used throughout Uganda. The level of each health center determines the medical support and capacity for testing and treatment that it has.

Dorothy and her team complemented this work by collaborating with another Marconi Society project in Guatemala that focuses on creating tools and techniques to assess Internet connection capabilities in remote and rural areas. This collaboration helps the Makerere students assess the connectivity capabilities for telemedicine at various clinics.

What We Have Learned at Makerere

In addition to the specific findings about the solutions noted above, Dr. Okello and her students have discovered some organizational and logistical truths that drive program success:

  • Summer time is the right time. At Makerere, it is better to run the program during the summer than during the academic school year. Since this is an extra-curricular activity, not a scheduled class, work on the Celestini Project slows down significantly during the school year due to other academic commitments. Engaging third year students during the summer also lets them identify a program-related capstone project for their senior year. They work on the capstone project during the school year, giving them adequate time to design a strong solution.
  • User perspectives are critical for designing the right solutions. The design-thinking workshops are imperative in helping students create solutions that solve the users’ problems, rather than cool designs that may not hit the sweet spot. As students experience these interactions, user-inspired design becomes a standard part of their problem-solving toolkit.
  • Faculty flexibility is key for success. “We are fortunate to have a lot of latitude to introduce new programs,” Dr. Okello said. “When faculty are introducing a new idea, they need time and a place to experiment before they turn it into a formal program or coursework.” Success of the program depends on the latitude that faculty has for introducing new initiatives without significant administrative overhead.
  • International collaborations are motivators. Students and faculty enjoy getting the perspectives of and leveraging the research of people in other countries who are doing complimentary work. Collaborating with the lead for the Marconi Society-sponsored project in Guatemala is a highlight for faculty and students since they get to learn new approaches to solving problems in other developing countries.
  • Leveraging the program contributions with other departments will ideally drive more university support.  By sharing information about the Celestini Program and the equipment and resources that netLabs!UG has gotten from its participation, Okello is working with colleagues to leverage these new resources to serve other parts of Makerere. Leveraging the resource from the program will ideally drive participating universities to put more resource behind it.

We look forward to another year of learning with netLabs!UG and Makerere students so that together we can reduce maternal mortality in Uganda and lay the foundation to scale the impacts of the Celestini Program.

Doubling Wireless Bandwidth: Just the Tip of the Iceberg

By Negar Reiskarimian and Paula Reinman

“I have a great respect for incremental improvement, and I’ve done that sort of thing in my life, but I’ve always been attracted to the more revolutionary changes … because they’re harder. They’re much more stressful emotionally. And you usually go through a period where everybody tells you that you’ve completely failed.”

— Steve Jobs

The Marconi Society stands for creativity, vision and change that benefits humankind.  Marconi Fellows have pioneered the Internet, wireless communications, security, GPS and other technologies that make today’s connected world possible.

The Marconi Society Young Scholars (nominations are open now!) show the potential to develop the next generation of ground-breaking discoveries.

So when 2017 Young Scholar Negar Reiskarimian decided what to focus on, she was not satisfied with developing integrated circuits that performed incrementally better. She wanted to find a fundamentally different way to design new circuits to power upcoming 5G and IoT applications.

Direction From the Intersection

As tomorrow’s applications diverge – from uses that will require more data and lower latency to support technologies such as the next generation of wireless communication systems (5G), augmented reality (AR) and virtual reality (VR) to Internet of Things (IoT) applications demanding lower power, smaller sizes and flexibility to tag everything – design ideas can come from many places across the scientific and engineering spectrum.  Negar’s research focuses on the intersection of fundamental physics, applied electromagnetics and nanoelectronics to design new integrated circuits (ICs) for these applications.

One key concept in overcoming current network design constraints is to create “full duplex” communications, enabling transmitters and receivers to function at the same time on the same frequency band. Full duplex capability doubles the capacity on existing networks and opens the door for 5G applications (click here for more information about full duplex and the ICs being developed at Columbia’s CoSMIC Lab).

Most of the prior full duplex demonstrations either used antenna pairs or nonreciprocal magnetic circulators to route the transmit and receive signals, which exist concurrently at the same frequency band. This is not the most efficient approach in terms of the area and cost of the overall system.

Negar’s challenge was to achieve non-reciprocal behavior without magnets.  She investigated prior approaches using modulation of material properties.  However these attempts could not meet the stringent requirements of full-duplex systems.

Combining her skills in IC design with her knowledge of the prior art, Negar created a different approach and a new modulation scheme based on switching transistors that is on a much smaller scale than anything that has been done before.  The intersection of this newly developed concept and CMOS platform produced a totally new type of design.

Beyond Doubling Capacity

While much of the initial interest around this innovation stems from its ability to double wireless network capacity, Negar believes that there are more fundamental and far-reaching applications for her work.

There are a number of government-funded programs to investigate and enhance the performance of nonreciprocal devices. DARPA’s SPAR (Signal Processing at Radio Frequency) program includes various groups working on nonreciprocal devices and circulators designed for high-performance wireless communication systems.  Negar and her team at CoSMIC Lab also have funding from the NSF to continue creating compact and low cost nonreciprocal devices that will support tomorrow’s applications.

Many of applications that seem to be right on the horizon will benefit from breaking reciprocity in a light, inexpensive and scalable way.  Most optical and electrical systems require nonreciprocal devices for protection against reflections. Self-driving cars need nonreciprocal devices in their radar systems to see objects and sense where they are on the road.

At a broader level, there is more that can be done with spatial and temporal modulation schemes beyond achieving nonreciprocity.  Since it is difficult to analyze and build these time varying systems, this work is just the beginning of endless opportunities.

New Rules for Building Game-Changing Technology

Solving the complex problems that will enable the next generation of connected communications and applications is an interdisciplinary sport.  Researchers need to look at how expertise in different fields and connecting ideas from these domains will result in new approaches to solve these problems. Scientists and engineers need to understand enough about the work and language of adjacent spaces to identify the promise of relevant ideas from other disciplines.

As basic scientific research continues to power consumer and business experiences, we will see more recognition of this research.  The 2017 Forbes 30 Under 30 List included two Marconi Society Young Scholars, Negar Reiskarimian and Dinesh Bharadia, who are creating big impact through fundamental research.

We hope to see more of this kind of innovation in the 2018 Young Scholar nominees.  Please click here for more information on nominating a deserving young innovator.