Skip to Content

CSE News

  • CSE Computer Security Startup Starts Selling Toolkit

    A startup out of the CSE department is flexing its muscles. This week, Tortuga Logic's co-founder and CEO, CSE alumnus Jason Oberg (PhD '13) announced immediate availability of the company's toolkit to transform the way hardware designers and system architects test the security of hardware designs.

    "The semiconductor industry needs to redirect its attention from only analyzing software vulnerabilities to identifying ways to detect security issues in hardware designs," said Oberg (at right) in a news release. "As more and more devices are designed to be Internet-enabled, the more we need to be concerned about hardware security. Hackers are focusing now on hardware." Tortuga Logic is a pioneer in the so-called design-for-security market. Its new toolkit, called Prospect, is able to uncover hidden bugs and prove the absence of vulnerabilities in hardware designs, thus minimizing security breaches in hardware and systems by automating the process of verifying security properties. Oberg's co-founders in Tortuga Logic include CSE Prof. Ryan Kastner and UC Santa Barbara Prof. Tim Sherwood, as well as Jonathan Valamehr, who is now the company's CFO and chief operating officer. This week's announcement put Tortuga Logic one step closer to graduating from the EvoNexus incubator, where it has been headquartered since late last year. The company should now be able to start adding commercial sales to a business that until now relied primarily on angel investors and NSF Small Business Innovation Research funding. In mid-April, Tortuga Logic was one of 16 startups participating in EvoNexus' Demo Day, where 350 attendees saw the company's demonstration of how their software analyzes the security properties in hardware designs to prevent cyber-security breaches. The company expects growing demand for its products with the advent of the Internet of Things. "Attacks have already been demonstrated on embedded devices such as pace makers, automobiles, baby monitors, and even refrigerators," Oberg told Chip Design magazine in late April. "Most companies are trying to solve this problem purely with software security, but this is a constant cat-and-mouse game we cannot win. As IoT grows, we are seeing more software being pushed down into hardware and our modern chipsets are growing in complexity. This is driving attackers to begin focusing on hardware and, without ensuring our chipsets are built in a secure manner, these attackers will continue to succeed.”

  • Students Put Finishing Touches to 3D, Multiplayer Networked Games

    Mark June 5th on your calendar. That's when five teams of students taking the CSE 125 capstone course on software system design and implementation -- better known as "the videogame course" -- will present their final projects and give audience members an opportunity to play the 3D, networked and multiplayer games in real time.

    Friday, June 5, 2015
    4pm - 5:30pm
    Calit2 Auditorium, Atkinson Hall

    The course has been taught since 2001, making this the 15th class to reach the finish line and produce their games from scratch. The final presentations in the Calit2 Auditorium are usually a standing-room-only celebration, as students from across campus flock to see how the teams  spent their 10 weeks (for many, the hardest 10 weeks of their entire undergraduate careers at UC San Diego, even though they also had fun). While creating a game adds excitement and motivation, CSE Prof. Geoffrey Voelker told students at the outset that "by the end of the course, you'll hopefully realize that what you learned in doing the project will apply to any large software project that is distributed, has performance constraints, has real-time constraints, has actual users other than developers, and so on." In short, CSE 125 gives seniors an opportunity to apply everything they should have learned in their major.

    Each team is usually broken into sub-teams to handle different parts of the challenge. For example, for Team 5, which is developing a "3D, multiplayer, player-vs.-player (PvP), team-based, territorial world conquer game," the entire six-person team is working on game play, while sub-teams are handling graphics (Robert Maloney and Sanjana Agarwal), networking (Kyle Parkinson and Benigno Baclig), as well as controls and art (Mimi Liu and Beth Yue Shi). While most of the games are first-person shooter games (including one where you kill your opponent by shooting bananas), there is also Battle Blocks, a sandbox building game where the player builds and customizes a battle robot to fight other bots, and a PvP fantasy combat game where each team of mechs (vampires/crusaders) is given a certain objective to complete in order to win each round.

    For the final presentations, which are open to the public and recorded for later on-demand viewing on the Internet, each team will demo their game, which must be played by four players drawn from the team and from the audience. A written project report is due at the end of finals week, but the final test is the June 5 demo. "The last thing you want," Voelker warned, "is a blue screen of death" when it's time to start playing the game in front of a real audience. 

  • 2015 Chancellor's Dissertation Medal Honors CSE Alumna

    CSE alumna Sarah Meiklejohn received her Ph.D. in computer science in 2014, and soon after accepted a faculty position at University College London in the UK, where she is a lecturer and assistant professor in the computer science as well as security-and-crime departments. For a researcher with broad interests in security and cryptography, being in London also offered the opportunity to continue her research on the international stage that she began at UC San Diego under CSE professors Mihir Bellare and Stefan Savage.

    Now comes word that Meiklejohn (at left) has been singled out for her dissertation, "Flexible Models for Secure Systems", as the recipient of the 2015 Chancellor's Dissertation Medal. As Bellare noted in recommending Meiklejohn for the medal, the results of her thesis "have shaped government policy", with the methods she proposed being used to track real-world cyber criminals. Regarding intellectual and technical depth, "the thesis introduces an innovative new experimental technique to track Bitcoins that was used not only to obtain the thesis results, but is now used as a key forensic tool by law enforcement," said Bellare, who also noted that Meiklejohn did it all herself: "She alone conceived the idea and methods and pushed it through from algorithms to reality. In my 20 years of experience at UCSD, I would say that a thesis with one of the above elements is rare. To have all three in the same thesis is unique and extraordinary."   

    In her dissertation, Meiklejohn explored how computer systems that seemed secure when they were launched then fell short because adversaries had capabilities that were not taken into account when the system was designed. "This gap between the abstract protocol and the deployed system leads to security concerns that ultimately impact every user of the system," she wrote. Meiklejohn went on to "model sophisticated attacks on the security of a system, such as side channels and fault injection, and then how to design flexible cryptosystems that can tolerate such attacks."  The thesis also looked at what she calls "the more benign scenario" where users cede some of their own privacy. Her landmark findings about user anonymity (or the lack of it) in the Bitcoin network -- which made headlines around the world while Meiklejohn was still at UC San Diego -- showed that she could combine publicly available information with minimal data gathered by hand to prove that the average Bitcoin user "is experiencing a fairly low level of anonymity, making Bitcoin ultimately unattractive for criminal activity such as money laundering" (even though it has been assumed that trading Bitcoin would not be visible to law enforcement authorities).

  • Gordon Center Stages Sixth Annual Think Tank

    This Friday and Saturday, the Gordon Engineering Leadership Center is hosting another of its "Think Tank" symposium series, this one focused on digital health.

    Think Tank: The Future of Digital Health
    May 15-16, 2015
    Friday 8am-5pm; Saturday 8am-2pm
    Qualcomm Conference Center
    Jacobs Hall
    Registration required

    Digital medicine is revolutionizing the way patients manage healthcare. This Think Tank will provide a comprehensive overview of digital health as well as unique perspectives from expert speakers from all facets of healthcare. The goal is to brainstorm the implications and uses of an emerging aspect of engineering, and past Think Tanks focused on topics as diverse as cyber security, cloud computing, and the energy crisis. This is the sixth annual Think Tank, and organizer Neil Gandhi explained that "improving data analysis to diagnosis, technology will have a sure impact on healthcare." The recent alumnus from Bioengineering also said that digital health is about applying what we know how to do with mobile devices, connectivity and big-data analytics to improve access to the healthcare system. Added Gandhi: "“New wireless devices paired with smart computing is enabling better patient care, but there are still challenges in integrating technology into the clinic and changing the way we already do things.” Speakers will address topics including big data, healthcare analytics, healthcare entrepreneurship, preventative medicine, mobile and global health, population health, emerging wireless technologies, and more. There will also be problem solving sessions, where small teams will have the opportunity to present their solutions to healthcare challenges.

  • Register to Participate in Big Pixel Hackathon

    Imagine being able to count every car in every city, monitor illegal elephant poaching from your couch, or measure the total area of solar panels on the planet. Geospatial big data is an emerging industry, with nearly a billion dollars in cash flowing into satellite imagery and analytics startups in just the last three years. The question on everyone's mind is: In all those pixels, what will we see? 

    Big Pixel Hackathon
    Saturday, May 23, 2015
    10am - 5pm
    Calit2 Theater, Atkinson Hall
    Register!

    On May 23, the Big Pixel Initiative at UC San Diego, a collaboration of the Qualcomm Institute and the newly-rechristened School of Global Policy and Strategy (GPS, previously IR/PS), will stage a day-long hackathon featuring three $500 cash prizes and opportunities to land mini-grants worth between $1,000 and $3,000 each for ongoing projects in geospatial big data. So far roughly 40 graduate and undergraduate students have registered to participate in the hackathon, with an expected top number of 75 students divided into teams that will have access to earth data sets supplied free of charge from the DigitalGlobe Basemap, the largest private collection of high-resolution satellite and aerial images (pictured) of the earth. Participants are invited to build apps that make unique use of the big pixel data.

    The Big Pixel Initiative is developing geospatial capacity to address our world's greatest challenges at scale, and organizers are looking to students for smart, early-stage multidisciplinary ideas. While coding is not required, early proof-of-concept work will increase competitiveness. Students are encouraged to bring their own laptops and to use the software platform of their choosing. Support stations will be provided for students to tap into industry-standard GIS software (ArcGIS, etc.). 

    The Initiative has partnered with the Digital Globe Foundation to grow a living, learning laboratory related to everything spatial, to investigate and design best practices in geospatial data visualization user experience interfaces, and design techniques for scientific discovery and decision-making.

  • CSE Gives Transfer Students a Head Start

    This summer CSE is one of several departments that will launch a new program for incoming transfer students arriving in Fall 2015. The Summer Academy will run from August 2 through September 5, and only transfer students majoring in computer science and engineering, biological sciences, electrical and computer engineering, or mathematics are eligible to apply to the five-week residential program on the UC San Diego campus. The program is partly modeled on the very successful Summer Program for Incoming Students (SPIS) launched two years ago to give incoming freshmen (pictured below) an opportunity to ease into campus life and get a strong foundation in computer science and programming prior to the start of the regular academic year.

    Transfer students accepted for the Summer Academy will earn eight units between two gateway courses to upper-division to "set a strong academic foundation leading to success at UC San Diego." Students will benefit from heavy academic advising and career mentoring, while developing strong ties with faculty and staff. The overall cost of the residential program will be primarily covered by the university itself, to the tune of $3,500 per student. The balance of costs, which include tuition, fees, educational programs, housing, meals, books and supplies, is $2,500, to be borne by students and their families (although financial aid is also available to students who qualify and submit an application by the June 1 deadline). Organizers point out that students accepted into the Summer Academy program are likely to graduate sooner, which could in turn offset some of the upfront cost of the program.

    Again, the deadline for incoming transfer students to apply for Summer Academy 2015 is June 1, and the same deadline applies to requests for financial aid under the 2014-15 FAFSA or California Dream Act. (Students who have already completed a financial aid application for 2014-15 should add UC San Diego's school code, 001317.)

  • CSE's Smarr Called 'Unlikely Hero' of Global Movement

    Calit2 Director and CSE Prof. Larry Smarr made the front page of The Washington Post. The May 9 article, “The Human Upgrade: The Revolution Will Be Digitized” by technology writer Ariana Cha, explores the movement to quantify consumers’ health and lifestyles, spearheaded by a flood of wearable devices such as the FitBit and Apple Watch.

    The article dubs Smarr “the unlikely hero of a global movement among ordinary people to ‘quantify’ themselves” using an estimated 211 million wearable monitoring devices to track their own health statistics (68 million devices shipped this year alone) – and using the knowledge to approach their physicians in new ways. “From the instant he wakes up each morning, through his workday and into the night, the essence of Larry Smarr is captured by a series of numbers: a resting heart rate of 40 beats per minute, a blood pressure of 130/70, a stress level of 2 percent, 191 pounds, 8,000 steps taken, 15 floors climbed, 8 hours of sleep,” is how Cha opens her feature article, adding: “Smarr, an astrophysicist and computer scientist, could be the world’s most self-measured man. For nearly 15 years, the professor at the University of California at San Diego has been obsessed with what he describes as the most complicated subject he has ever experimented on: his own body.”  Smarr monitors more than 150 parameters related to his health and activity, and he compares his self-monitoring with the way many Americans monitor their own cars. "We know exactly how much gas we have, the engine temperature, how fast we are going," he is quoted as saying. "What I'm doing is creating a dashboard for my body." (Pictured: Smarr with some of the 150 parameters depicted on a wall display in Calit2's Qualcomm Institute; photo by Earnie Grafton for The Washington Post.)

    Read the original article in The Washington Post.

  • Knight, Smarr Raise Profile of Microbiome Research

    Two computer scientists at UC San Diego are raising the profile of research on the human microbiome -- the 100 trillion human cells in and on each human body, of which only 10 trillion are human. The correct balance of microbes can keep pathogens in check and regulate the immune system, and microbes are essential in food digestion and synthesizing vitamins.

    Rob Knight (left), a CSE faculty-affiliate and professor of pediatrics in the School of Medicine, is making headlines again, reporting that microbial communities on adult skin are on average more similar to those of their own dogs than to other dogs. "With a fair amount of precision," Knight said, "we could pick your dog out of a crowd based solely on microbiomes." The study raises questions about how the family dog may affect an individual's health, given that a human and their dog share similar microbial populations. For instance, Knight says there is mounting evidence that living with a dog in infancy lowers a child's risk of developing asthma and allergies. The professor co-founded the American Gut Project, a crowdsourcing initiative to get people to contribute mouth, skin or gut samples of their own, their family members, and their dogs for microbiome sequencing. Participants get to find out what microbe populations they share with others, and researchers get a growing database on which to base future studies of the microbiome and the influence that having a dog and other factors may have on human health. Knight's lab is also sequencing microbiomes submitted to a University of Arizona study focused specifically on dogs and human health. "The idea of combining animal, human and environmental health, and seeing the whole picture through the lens of the microbes that we share, is an increasing direction for research," Knight told the San Diego Union-Tribune.

    Meanwhile, another CSE professor is involved in an effort to make microbiome sequencing more accessible to researchers. The company uBiome makes a toolkit for researchers, and it just announced a global academic research consortium to study microbiomes and health. Part of the challenge involves computing and 'quantified health' technologies, so it was no surprise when the company announced this week that CSE Prof. Larry Smarr (right) had been appointed to the scientific advisory board of uBiome in connection with the launch of the global consortium "to bring together talented researchers worldwide... to complement, not replace or compete with, efforts like the IHMC and NIH Human Microbiome Project." Membership is open to researchers worldwide, with beneits that could include up to $100,000 worth of microbiome kits to support exceptional research proposals, especially in underrepresented areas. As uBiome noted in its press release announcing the consortium and Smarr's involvement, "the latest research suggests that the correct balance of microbes serves to keep potential pathogens in check and regulate the immune system." For its part, uBiome provides participants with a catalog of their sequenced microbiomes through a self-collected sample that is then processed in uBiome's San Francisco lab. Joining CSE's Smarr on the uBiome advisory board announced May 7 are UCSF professor and MacArthur Genius Award winner Joseph DeRisi, and Jonathan Eisen, a professor of microbiology at UC Davis.

  • CSE Students Invited to Apply for SDSC Summer Institute

    The San Diego Supercomputer Center (SDSC) has announced the topic of its 2015 Summer Institute, which runs from August 10-14: High Performance Computing for the Long Tail of Science. The Summer Institute is a week-long education and training program in HPC and data-intensive computing. Priority applications are due by Friday, May 29.

    The first half will consist of plenary sessions covering essential skills including data management, running jobs on SDSC resources, reproducibility, database systems, characteristics of big data, and techniques for turning data into knowledge, software version control, and making effective use of hardware. This will be followed by parallel sessions to dive deeper into specialized materials relevant to research projects. Parallel session will cover topics including SPARK, parallel computing, performance optimization, predictive analytics, scalable data management, visualization, workflow management, GPUs/CUDA, and Python for scientific computing. SDSC says the program "will deploy a flexible format designed to help attendees get the most out of their week." Participants will also get the opportunity to do hands-on exercises on Comet (SDSC's newest supercomputer), and the data-intensive Gordon system.

    The Summer Institute is encouraging applicants interested in data science and computational science, and especially, current and potential users of SDSC's data-intensive resources. Some familiarity with UNIX/Linux environments is recommended, along with some programming experience in C/C++, Fortran, Java, R, Python, Perl, MATLAB or other languages. Applicants already dealing with data science and/or computational science problems are especially encouraged to apply. SDSC opens up the Summer Institute to people outside of San Diego, but participants are encouraged to apply early, because the number of available room-and-board scholarships is limited. Priority applications are due Friday, May 29 (for notification by June 5 whether you've been admitted). Applications after May 29 will be considered  based on experience and availability. Applicants must be in one of three categories: academic faculty or staff; graduate and undergraduate students or post-doctoral researchers; or industry.

  • Computer Vision + Brain-Computer Interface = Faster Mine Detection

    Computer scientists in UC San Diego's Jacobs School of Engineering have combined sophisticated computer vision algorithms and a brain-computer interface to find mines in sonar images of the ocean floor. The study shows that the new method speeds detection up considerably, when compared to existing methods—mainly visual inspection by a mine detection expert.

    “Computer vision and human vision each have their specific strengths, which combine to work well together,” said CSE Prof. Ryan Kastner (pictured at right). “For instance, computers are very good at finding subtle, but mathematically precise patterns while people have the ability to reason about things in a more holistic manner, to see the big picture. We show here that there is great potential to combine these approaches to improve performance.”

    Researchers worked with the U.S. Navy’s Space and Naval Warfare Systems Center Pacific (SSC Pacific) in San Diego to collect a dataset of 450 sonar images containing 150 inert, bright-orange mines placed in test fields in San Diego Bay. An image dataset was collected with an underwater vehicle equipped with sonar. In addition, researchers trained their computer vision algorithms on a data set of 975 images of mine-like objects.

    In the study, researchers first showed six subjects a complete dataset, before it had been screened by computer vision algorithms. Then they ran the image dataset through mine-detection computer vision algorithms they developed, which flagged images that most likely included mines. They then showed the results to subjects outfitted with an electroencephalogram (EEG) system, programmed to detect brain activity that showed subjects reacted to an image because it contained a salient feature—likely a mine. Subjects detected mines much faster when the images had already been processed by the algorithms. Computer scientists published their results recently in the IEEE Journal of Oceanic Engineering.

    The algorithms are what’s known as a series of classifiers, working in succession to improve speed and accuracy. The classifiers are designed to capture changes in pixel intensity between neighboring regions of an image. The system’s goal is to detect 99.5 percent of true positives and only generate 50 percent of false positives during each pass through a classifier. As a result, true positives remain high, while false positives decrease with each pass.

    Researchers took several versions of the dataset generated by the classifier and ran it by six subjects outfitted with the EEG gear, which had been first calibrated for each subject. It turns out that subjects performed best on the data set containing the most conservative results generated by the computer vision algorithms. They sifted through a total of 3,400 image chips sized at 100 by 50 pixels. Each chip was shown to the subject for only 1/5 of a second (0.2 seconds) —just enough for the EEG-related algorithms to determine whether subject’s brain signals showed that they saw anything of interest.

    All subjects performed better than when shown the full set of images without the benefit of prescreening by computer vision algorithms. Some subjects also performed better than the computer vision algorithms on their own.

    “Human perception can do things that we can’t come close to doing with computer vision,” said Chris Barngrover, who earned a computer science Ph.D. in Kastner’s research group and is currently working at SSC Pacific. “But computer vision doesn’t get tired or stressed. So it seemed natural for us to combine the two.” CSE Ph.D. student Alric Althoff also participated in the study.



about seo