News

David Patterson is leading one of Google's most crucial projects

Prof. Emeritus David Patterson is profiled in a CNBC article which describes how he postponed retirement to conduct research at Google into the Tensor Processing Unit (TPU), an ambitious new chip that's designed to run at least 10 times faster than today's processors and is sophisticated enough to handle the intensive computations required for artificial intelligence.  Without it, it is estimated that Google would have to double its data centers to support even a limited amount of voice processing.  Prof. Patterson described his work on the TPU when he returned to Berkeley as a Colloquium speaker on May 3rd.

Jan Rabaey and Pieter Abbeel named in the top 5 of the 2017 top 50 tech pioneers by the Belgian financial times

Professors Jan Rabaey and Pieter Abbeel were named in the top 5 of the 2017 top 50 pioneers by the “De Tijd” (translation buttons provided above the article), the Belgian equivalent of the Financial Times. Prof. Rabaey is currently the scientific co-director of the Berkeley Wireless Research Center (BWRC) as well as the director of the FCRP Multiscale Systems Research Center (MuSyC), and is involved with the Donald O. Pederson Center for Design Automation (DOP)SWARM Lab,  CITRIS People and Robots (CPAR)TerraSwarm Research Center, and the  Center for Neural Engineering & Prostheses (CNEP). His research interests include ultra-low energy wireless exploring the boundaries of ultra-low energy design and the design of microscopic systems, including all components from energy sources, conversion and storage, interfaces, digital and mixed signal. Prof. Abbeel is currently a member of the steering committee of the Berkeley Artificial Intelligence Research Center (BAIR) and is involved with the Center for Human Compatible Artificial Intelligence (CHCAI)Berkeley Vision and Learning Center (BVLC)Center for Automation and Learning for Medical Robotics (Cal-MR) and CITRIS People and Robots (CPAR). His current research area is primarily studying deep learning for robotics, where learning could be from demonstrations (apprenticeship learning) or through the robot's own trial and error (reinforcement learning). Targeted application domains include autonomous manipulation, flight, locomotion and driving.

Jitendra Malik recipient of the ACM and AAAI Allen Newell Award

Prof. Jitendra Malik has been named recipient of the Association for Computing Machinery (ACM) and Association for the Advancement of Artificial Intelligence (AAAI) Allen Newell Award. The Allen Newell award is presented to an individual for career contributions that have breadth within computer science, or that bridge computer science and other disciplines. Prof. Malik's research has addressed several important problems in computer vision: how to characterize contours in images, how to segment images, and how to represent shape for feature matching.  He also was a leader in evaluation methods through the creation of the Berkeley segmentation dataset, using human segmentations to evaluate the correctness of the algorithmic segmentations.  He pioneered the use of normalized cuts, anisotropic diffusion, high dynamic range imaging, shape context and the use of graph theory for low-level to mid-level computer vision problems.  In computer graphics, his research showed how digital photographs and user-guided photogrammetry can be used to synthesize highly photorealistic computer-generated architectural scenes.  He also has made important contributions to computational models of human texture perception including segmentation, shape from texture, and intrinsic image computation.

Berkeley CS faculty among the most influential in their fields

U.C. Berkeley has the top ten most AMiner Most Influential Scholar Award winners across all fields of computer science in 2016 and the top five most award winners in the fields of Computer Vision, Database, Machine Learning, Multimedia, Security, Computer Networking, and System.  The 28 CS faculty members included in the rankings were among the 100 most-cited authors in 12 of the 15 research areas evaluated. Two were among the 100 most-cited authors in 3 different areas each: Scott Shenker ranked #1 in Computer Networking, #51 in System, and #99 in Theory; and Trevor Darrell ranked #8 in Mulitmedia, #18 in Computer Vision, and #100 in Machine Learning.  Out of the 700,000 researchers indexed, only 16 appeared on three or more area top 100 lists.  See a more detailed breakdown of our influential faculty scholars.

Alyosha Efros has won the 2016 ACM Prize in Computing

Professor Alexei (Alyosha) Efros has won the 2016 Association of Computing Machinery (ACM) Prize in Computing, formerly known as the ACM-Infosys Foundation Award. This award recognizes early-to-mid-career contributions that have fundamental impact and broad implications. Prof. Efros was cited for groundbreaking data-driven approaches to computer graphics and computer vision and is a pioneer in combining the power of huge image datasets drawn from the Internet with machine learning algorithms to foster powerful image transformations and valuable research findings. He has also made fundamental contributions in texture synthesis, a technique that ushered in new horizons in computer graphics and is widely used in the film industry. ACM Prize recipients are invited to participate in the Heidelberg Laureate Forum, an annual networking event that brings together young researchers from around the world with recipients of the ACM A.M. Turing Award (computer science), the Abel Prize (mathematics), the Fields Medal (mathematics), and the Nevanlinna Prize (mathematics).

Certificate in Design Innovation is launched

A new certificate program, the Berkeley Certificate in Design Innovation (BCDI),  is the result of a cross-disciplinary, cross-departmental partnership between the College of Environmental Design, the College of Engineering, the Haas School of Business and the College of Letters and Science’s Arts and Humanities Division.  It offers all undergraduates at Cal an opportunity to foster a critical mindset, collaboratively define problems, and develop the technical proficiency to innovate broadly outside of their major.  CS Associate Professor and BCDI advisor Eric Paulos says “I can’t wait to see what this cross pollination of design methods, materials, tools and most of all people will bring to not just UC Berkeley but to our communities – from the local to the global.”  BCDI will be hosting an open house event on Friday, April 21 at noon at Jacobs Hall for prospective students to learn more about the Certificate, meet faculty members involved, and hear from guest speakers and UC Berkeley alumni.

Jitendra Malik will be a keynote speaker at the 2017 Embedded Vision Summit

EECS Chair and CS Prof. Jitendra Malik will discuss Deep Visual Understanding from Deep Learning as one of the keynote speakers at the Embedded Vision Summit on May 2, 2017.  The summit is the only event focused exclusively on the technologies, hardware, and software that bring visual intelligence to products.  This year, "It's all about deployable computer vision and deep learning" and will feature more than 90 expert presenters in 4 conference tracks over three days.

Kathy Yelick elected to the American Academy of Arts and Sciences

Prof. Katherine Yelick has been elected to the American Academy of Arts and Sciences. This organization has been serving the nation as a champion of scholarship, civil dialogue and useful knowledge since 1780. The Academy convenes leaders from the academic, business, and government sectors to address critical challenges facing our global society. Kathy joins a long list of distinguished members, going back to Ben Franklin, Alexander Graham Bell, and most recently our own Scott Shenker in 2016. For a complete list of EECS members elected to the academy, see EECS Faculty Awards/American Academy of Arts and Sciences.

Google TPUs are built for inference

CS Prof. Emeritus David Patterson co-authored and presented a report on Tensor Processing Units (TPUs) at a regional seminar of the National Academy of Engineering, held at the Computer History Museum in Menlo Park on April 5, 2017.   TPUs, which have been deployed in Google datacenters since 2015, are printed-circuit cards which are inserted into existing servers and act as co-processors tailored for neural-network calculations.  Prof. Patterson says that TPUs are "an order of magnitude faster than contemporary CPUs and GPUs" with an even larger relative performance per watt.  According to an article for the IEEE Spectrum, TPUs are "built for doing inference," having hardware that operates on 8-bit integers rather than the higher-precision floating-point numbers used in CPUs and GPUs.