Two UMBC alumni who now work at Google will present a Google information session and two technical talks this coming Wednesday afternoon in the Library Gallery. It's a chance for both undergraduate and graduate students to learn about internship and full-time opportunities at Google. The Googlers are:

• Blazej Bulka is a currently a Senior Software Engineer at Google NYC. He received a Ph.D. in Computer Science from UMBC in 2009.
• Robert Banz is a Site Reliability Manager at Google. He received a BS in Computer Science from UMBC in 1995 and worked for UMBC's Department of Information Technology for many years before joining Google.

There will be an opportunity to network and ask questions between talks and an informal happy hour after the talks starting at 5:00pm at Flat Tuesdays in the Commons. Faculty, staff and frinds are welcome to attend the talks and happy hour.

• Noon-1:00pm Google Culture and Careers, Presented by Blazej Bulka and Robert Banz

  Google culture and mission presented by two UMBC graduates who work at Google. Brief history of Google, products and projects, and first-hand relation of how it is to work for Google. The second half of the talk will describe various career opportunities (both for full-time positions and internship), including information about the interview process.

• 2:00-3:00pm Engineering Reliability at Google, Presented by Robert Banz

  Keeping Google services up and running efficiently, reliably, and at scale.

• 4:00-5:00pm Research and Work at Google: Graduate Student Opportunities, Presented by Blazej Bulka

  How to do research and utilize skills learned in graduate school at Google.

• 5:00- Happy Hour at Flat Tuesdays, The Commons

Let Google know you are interested in opportunities and plan to attend some of the sessions by filling out this form.

A study by CSEE professor Cynthia Matuszek on gender bias in online images associated with a variety of occupations has recently received a lot of attention. The study, which was carried out with former University of Washington colleagues Matthew Kay and Sean Munson, resulted in a paper to be presented at the ACM CHI Conference on Human Factors in Computing Systems in Seoul, Korea in mid-April.

The paper, Unequal Representation and Gender Stereotypes in Image Search Results for Occupations, was recognized as a best paper of the 2015 ACM CHI Conference. This conference is considered the most prestigious in the field of human–computer interaction, and is one of the top ranked computer science conferences. 

"This project was a joint effort between myself and the two other authors," Dr. Matuszek explains. "The project was originally developed after I saw a presentation in which the images chosen to represent different professions were extremely one-sided and didn't seem representative.  Our primary goal was to learn more about the software design space: does gender in image search results affect whether people think the search results are good?  On the flip side, does showing different genders in different roles (that is, showing a male vs. a female nurse) affect how people think of those occupations?  You need to know the answers to questions like that before you can create, e.g., an image search engine that is really well and thoughtfully designed."

Not surprisingly, Dr. Matuszek's work has been noted by the popular press, with recent stories in the Washington Post Wonkblog, The Atlantic and many online news outlets. The Washington Post article quotes Dr. Matuszek on the genesis of the study.

In addition, many images retrieved by the web’s top search engine happen to be hyper-sexualized caricatures. Some female construction workers in midriff-baring flannel and jean shorts seem better dressed for a Halloween party than, say, a demolition site. (Researchers dubbed this the “sexy construction worker problem.”)

"It’s part of a cycle: How people perceive things affects the search results, which affect how people perceive things," said co-author Cynthia Matuszek, who now teaches computer ethics at the University of Maryland, Baltimore County. Matuszek recalls sitting in a robotics lecture last year at the University of Washington, where she earned her doctoral degree in computer science. A male colleague illustrated researchers in his Powerpoint presentation as “all guys, classic nerds,” she said. But a caretaker was shown in a slide as "a plump woman in her thirties who was wearing a pink suit." The stereotypes irked Matuszek, and she's not the only one wondering about the power of images.

About eight months ago, Matuszek and her colleagues at the University of Washington decided to test the power of popular image. They wanted to know if something as seemingly trivial as search results could sway someone’s perception of how many women work in a certain field, and whether they’re competent. The researchers surveyed 21 people — a pool too small to make any sweeping statement, Matuszek acknowledges, but big enough for a glimpse into our cultural psyche — starting with questions like: What percentage of construction workers are women? Do you believe the person in this photo is good at their job? Two weeks later, they followed up, prompting participants to sift through Google image results before answering the same inquiries. Responses changed after Google images were introduced, according to the study, which was published this week. Search results could determine 7 percent of a participant's subsequent opinion about the number of men and women in a particular field, the authors calculated. And a worker was, on average, deemed more competent if he or she fit into a gender stereotype.

As Adrienne LaFrance notes in a recent Atlantic article about Matuszek's study, "Google image searches don't just reflect the sad state of diversity in corporate leadership; they actually influence the ways in which people think about what it means to be a CEO."

The study concluded that "shifting the representation of gender in image search results can shift people’s perceptions about real-world distributions."[i]

Northrop Grumman Corporation announced an expansion of their cybersecurity work with UMBC to include research on health data analytics in partnership with the UMBC Center for Hybrid Multicore Productivity Research (CHMPR). CHMPR is an NSF-sponsored industry/university cooperative research center and consortium focused on addressing the productivity, performance and scalability issues in meeting the computational demands of its sponsors applications through the continuous evolution of multicore architectures and open source tools.

From the Northrop Grumman press release:

The UMBC-Northrop Grumman collaboration began with a cybersecurity initiative and is now leveraging the big data analytics used in cyber and applying the tools and technologies to health. This has enabled Northrop Grumman to dive into expansive health data on a variety of subjects, such as the impacts of cardiovascular disease as it relates to smoking, obesity and pharmaceutical drug use.

"CHMPR has provided greater clinical access to information, analytic techniques and natural language processing," said Amy Caro, vice president and general manager, health division, Northrop Grumman Information Systems. "Our efforts have yielded real-time results and are driving programs and practices that are helping to improve the health of large populations. As we move forward, our security capabilities and work with CHMPR will allow for continued large scale data mining to help inform breakthrough research and drive better health outcomes."

Northrop Grumman has been a member and supporter of CHMPR, since its inception and chairs the CHMPR Industry Advisory Board. Northrop Grumman and UMBC started their participation with CHMPR focusing on cyber to demonstrate a commitment to that mission and have since expanded that focus to include high performance computing and big data analytics in the health arena. Most recently, Northrop Grumman has provided initial funds to CHMPR's new North Carolina State University planning site.

"Our partnership with Northrop Grumman and our work in cyber and now health are huge steps forward in advancing critical research and driving better business and health results. We couldn't be prouder of these accomplishments and we look forward to continuing this great work," said Yelena Yesha, professor of computer science and CHMPR director, UMBC.

"Working with industry to advance research is a great model and Northrop Grumman and UMBC have been a key leaders and partners in building this industry, government and university consortium to advance issues like cyber security and health analytics," said Rita Rodrigues, program director, National Science Foundation.

Northop Gumman is collaborating with UMBC on several initiatives supporting cybersecurity education, training and technology. These include partnering with the bwtech@UMBC Cyber Incubator to form Cync, a program that nurtures innovative cybersecurity startups and helping UMBC launch its Cyber Scholars Program.

Dr. Rick Forno, CSEE’s Cybersecurity Graduate Program Director, was a presenter at the inaugural BSides Charm City technical security conference held at Howard County Community College in Columbia, MD over the weekend of 11-12 April 2015. His talk described what is needed to develop and operate viable high school and college student cyber-competition teams and framed around what he called the ‘POTS process’ — People, Organization, Technology, and Sustainability.

Several lessons learned from his four years overseeing the Maryland Cyber Challenge (MDC3) served as the basis for his presentation remarks, including the need to foster greater diversity and acceptance within cyber-competition teams — and by extension, the cybersecurity industry itself.

Baltimore City Teaching Residency – Final Deadline Extended to April 13

Baltimore City Teaching Residency trains UMBC graduates from a wide variety of majors and backgrounds to become exceptional, certified teachers for Baltimore City students in high-need communities.  It’s an honor and a tremendous responsibility, which is why we hold the highest standards of effectiveness of any teacher preparation program in America. 

Our 1-year program provides teacher training and certification while you work as a fulltime teacher, thus earning full salary and benefits.

Do you have what it takes to join this elite group of educators?

Click here – our final application deadline is Monday, April 13.  Have questions? Email anhthi.mouradov at tntp.org.

Our training program includes:

• 6-week pre-service training (Summer 2015, 5 days per week, fulltime hours)
• Weekly TNTP Academy training seminars (throughout the 2015-16 school year)
• Expert coaching to ensure your success throughout your first year teaching
• Extensive help and resources to secure your fulltime teaching position for the 2015-16 school year and beyond

Baltimore City Teaching Residency – Final Deadline Extended to April 13

Baltimore City Teaching Residency trains UMBC graduates from a wide variety of majors and backgrounds to become exceptional, certified teachers for Baltimore City students in high-need communities.  It’s an honor and a tremendous responsibility, which is why we hold the highest standards of effectiveness of any teacher preparation program in America. 

Our 1-year program provides teacher training and certification while you work as a fulltime teacher, thus earning full salary and benefits.

Do you have what it takes to join this elite group of educators?

Click here – our final application deadline is Monday, April 13.  Have questions? Email anhthi.mouradov at tntp.org.

Our training program includes:

• 6-week pre-service training (Summer 2015, 5 days per week, fulltime hours)
• Weekly TNTP Academy training seminars (throughout the 2015-16 school year)
• Expert coaching to ensure your success throughout your first year teaching
• Extensive help and resources to secure your fulltime teaching position for the 2015-16 school year and beyond

Baltimore City Teaching Residency – Final Deadline Extended to April 13

Baltimore City Teaching Residency trains UMBC graduates from a wide variety of majors and backgrounds to become exceptional, certified teachers for Baltimore City students in high-need communities.  It’s an honor and a tremendous responsibility, which is why we hold the highest standards of effectiveness of any teacher preparation program in America. 

Our 1-year program provides teacher training and certification while you work as a fulltime teacher, thus earning full salary and benefits.

Do you have what it takes to join this elite group of educators?

Click here – our final application deadline is Monday, April 13.  Have questions? Email anhthi.mouradov at tntp.org.

Our training program includes:

• 6-week pre-service training (Summer 2015, 5 days per week, fulltime hours)
• Weekly TNTP Academy training seminars (throughout the 2015-16 school year)
• Expert coaching to ensure your success throughout your first year teaching
• Extensive help and resources to secure your fulltime teaching position for the 2015-16 school year and beyond

UMBC's Professor Tim Oates has a column on the online TechCrunch site describing why we should Stop Fearing Artificial Intelligence. Professor Oates has 20 years of experience working with a wide range of AI technologies, including machine learning, robotics and natural language processing. In the piece, Dr. Oates explains that

"As yet another tech pioneer with no connection to artificial intelligence steps out to voice his fears about AI being catastrophic for the human race, I feel the need respond. … Conflating facts of technology's rapid progress with a Hollywood understanding of intelligent machines is provocative (honestly, it's a favorite in my most-loved science fiction books and movies), but this technology doesn't live in a Hollywood movie, it isn't HAL or Skynet, and it deserves a grounded, rational look.

and discusses some of the limitations of current intelligent systems like IBM's Watson. Like most AI researchers, he's a believer in Strong AI — the idea that there is no theoretical reason why a machine can not exhibit behavior as skillful and flexible as humans — but doubts the such machines will be neccessarily dangerous.

"But let's suppose, for a second, that an AI does learn to think intelligently outside its programming and that it’s become discontent. Would this superhuman intelligence inherently go nuclear, or would it likely just slack off a little at work or, in extreme cases, compose rap music in Latin? In a world filled with a nearly infinite number of things a thinking entity can do to placate itself, it's unlikely "destruction of humanity" will top any AI's list."

Stop Fearing Artificial Intelligence is a well written and thought provoking article.

PhD Dissertation Defense

Physical- and MAC-Layer Mechanisms for Increasing
Base-Station Anonymity in Wireless Ad-hoc Networks

John Ward

10:00am Monday, 13 April 2015, ITE 325b, UMBC

Wireless ad hoc networks have become valuable assets to both the commercial and military communities with applications ranging from industrial control on a factory floor to reconnaissance of a hostile border. In most applications, data sources forward information over multi-hop paths to a base station (BS). The BS not only serves as the data sink, but also provides other basic control and management features such as protocol synchronization, a gateway to other networks, and operator failure notifications, without which the network becomes dysfunctional. The critical role of the BS makes it a natural target for an adversary that desires to achieve the most impactful attack on the network. Even if an ad-hoc network employs conventional security mechanisms such as encryption and authentication, an adversary may apply traffic analysis techniques to identify the BS. This motivates a significant need for achieving improved BS anonymity to conceal its identity, role, and location. Traffic analysis countermeasures found in the literature have focused on the link and network layers and little attention has been given to the physical-layer (PHY) approaches.

In this dissertation, we address the challenges of BS anonymity by developing PHY and Medium Access Control (MAC) traffic analysis countermeasures. We first consider the limitations associated with evidence theory, the adversary’s primary traffic analysis attack model. We analyze the susceptibility of evidence theory to imperfect received signal strength (RSS) measurements. Next we extend the standard evidence theory approach to consider the contribution of acknowledgements to BS anonymity. We call this approach Acknowledgement Aware Evidence Theory (AAET) and describe the adversary’s initial survey phase which is based on a novel message correlation approach. Accurate synchronization is required by an ad-hoc network to employ PHY countermeasures; however, synchronization represents a fundamental design tradeoff, since the low-cost nodes and conservation of energy limit the achievable accuracy of synchronization within the network. We consider the impact of two popular synchronization protocols, Reference Broadcast Synchronization (RBS) and Timing-Synch Protocol for Sensor Networks (TPSN) on BS anonymity.

We develop a novel PHY BS anonymity-boosting approach based on distributed beamforming, which we call Distributed Beamforming protocol for increased BS ANonymity (DiBAN). DiBAN is a protocol that includes a cross-layer relay selection algorithm that a node employs to determine the most appropriate number of helper relays to participate in distributed beamforming at each hop. While effective at increasing BS anonymity, the overhead associated with DiBAN is in addition to any overhead already associated with existing ad hoc network protocols. We refine DiBAN using cross-layer design methodology to create an integrated distributed beamforming protocol called Link and Physical Cross-layer design for increased Anonymity (LiPCA) that increases BS anonymity with minor additional overhead by taking advantage of the underlying services that the MAC already provides. We use simulation to demonstrate the anonymity performance and energy-efficiency of our approaches and use the popular Sensor MAC (S-MAC) protocol as an exemplar MAC protocol. Results show that the cross-layer design of our LiPCA protocol achieves equivalent anonymity performance to DiBAN, but with an energy savings of approximately 35 percent.

Committee: Drs. Mohamed Younis (Chair), Charles Nicholas, Richard Forno, Ryan Robucci and Pedro Rodriguez (JHU/APL)

Ph.D. Dissertation Defense
Electrical Engineering

Simultaneous Polarization-Insensitive Phase-space
Trans-multiplexingand Wavelength Multicasting via
Cross-phase Modulation in a Photonic Crystal Fiber at 10 GBd

Brice Cannon

2:00pm Monday, 6 April 2015, ITE325b, UMBC

This thesis investigates the all-optical combination of amplitude and phase modulated signals into one unified multi-level phase modulated signal, utilizing the Kerr nonlinearity of cross-phase modulation (XPM). Predominantly, the first experimental demonstration of simultaneous polarization-insensitive phase-transmultiplexing and multicasting (PI-PTMM) will be discussed. The PI-PTMM operation combines the data of a single 10-Gbaud carrier-suppressed return-to-zero (CSRZ) on-off keyed (OOK) pump signal and 4×10-Gbaud return-to-zero (RZ) binary phase-shift keyed (BPSK) probe signals to generate 4×10-GBd RZ-quadrature phase-shift keyed (QPSK) signals utilizing a highly nonlinear, birefringent photonic crystal fiber (PCF). Since XPM is a highly polarization dependent nonlinearity, a polarization sensitivity reduction technique was used to alleviate the fluctuations due to the remotely generated signals’ unpredictable states of polarization (SOP). The measured amplified spontaneous emission (ASE) limited receiver sensitivity optical signal-to-noise ratio (OSNR) penalty of the PI-PTMM signal relative to the field-programmable gate array (FPGA) pre-coded RZ-DQPSK baseline at a forward-error correction (FEC) limit of 10-3 BER was ≈ 0.3 dB. In addition, the OSNR of the remotely generated CSRZ-OOK signal could be degraded to ≈ 29 dB/0.1nm, before the bit error rate (BER) performance of the PI-PTMM operation began to exponentially degrade. A 138-km dispersion-managed recirculating loop system with a 100-GHz, 13-channel mixed-format dense-wavelength-division multiplexed (DWDM) transmitter was constructed to investigate the effect of metro/long-haul transmission impairments. The PI-PTMM DQPSK and the FPGA pre-coded RZ-DQPSK baseline signals were transmitted 1,900 km and 2,400 km in the nonlinearity-limited transmission regime before reaching the 10-3 BER FEC limit. The relative reduction in transmission distance for the PI-PTMM signal was due to the additional transmitter impairments in the PCF that interact negatively with the transmission fiber.

Committee: Drs. Professor Gary M. Carter (Chair), William Astar, Anthony M. Johnson, Tinoosh Mohsenin, Thomas E. Murphy, Terrance L. Worchesky