News

Vasuki Narasimha Swamy and César Torres win Microsoft Research Dissertation Grants

EE graduate student Vasuki Narasimha Swamy (adviser: Anant Sahai) and CS graduate student César Torres (adviser: Eric Paulos) have won inaugural Microsoft Research Dissertation Grants.  These grants offer financial support to selected doctoral students from groups that are under-represented in the field of computing.  Of the 200 applicants, only 12 were chosen.  Vasuki's research topic is “Real-time Ultra-reliable Wireless Communication” and César's is “Hybrid Aesthetics – A New Media Framework for the Computational Design of Creative Materials, Tools, and Practices within Digital Fabrication.”

Edgar Solomonik wins Householder Prize for best dissertation in numerical linear algebra

Alumnus Edgar Solomonik (CS Ph.D. '14,  adviser: James Demmel) has won the Alston S. Householder Prize XX (2017)  for the best dissertation in numerical linear algebra.  The Householder prize, which is presented once every three years at a Symposium held in cooperation with the Society for Industrial and Applied Mathematics (SIAM) and the SIAM Activity Group on Linear Algebra, has two winners this year.  Edgar's dissertation,  titled "Provably Efficient Algorithms for Numerical Tensor Algebra," also won the EECS Department's David J. Sakrison Memorial Prize for truly outstanding research. Edgar is currently an Assistant Professor in the Department of Computer Science at the University of Illinois at Urbana-Champaign.

Dawn Tilbury named head of NSF Engineering Directorate

EECS alumna Dawn Tilbury (M.S.'92, Ph.D.'94) will lead investments in fundamental engineering research and education as the newly appointed head of the National Science Foundation's (NSF) Directorate for Engineering (ENG).  ENG supports engineering research and education critical to the nation's future and fosters innovations that benefit society.   It provides about 32 percent of the federal funding for fundamental research in engineering at academic institutions.  Dawn has been a Professor of Mechanical Engineering at the University of Michigan, focusing in the area of control systems, with a courtesy appointment in the EECS department.  She also formerly served as associate dean for research in UMich's College of Engineering.  "I welcome the opportunity to work with the engineering and scientific community to address the big challenges that face the nation and world today," she says.

Jose Carmena in IEEE article titled “Timeline: The Evolution of Assistive Technologies”.

Jose Carmena is mentioned in IEEE’s publication The Institute in an article titled “Timeline: The Evolution of Assistive Technologies”  (see timeline at the bottom of the article, year 2017). In celebration of The Institute's 40th anniversary this article highlights topics and technologies  over the past four decades that have applied electronics to significantly help people overcome disabilities. Professor Carmena is recognized for contributions to the neural basis of motor-skill learning and neuroprosthetic systems. His research program in neural engineering and systems neuroscience is aimed at understanding the neural basis of sensorimotor learning and control, and at building the science and engineering base that will allow the creation of reliable neuroprosthetic systems for the severely disabled. He is also co-chair of the IEEE Brain Initiative and Co-Director of the Center for Neural Engineering and Prostheses at UC Berkeley and UCSF.

Sculpted Light in the Brain

In an effort to gather scientists at the interface between neurosciences, optical engineering, and computer science, an all-day conference is being held on Friday, June 9, in Stanley Hall titled Sculpted Light in the Brain.   Participants are united in their mission to develop technologies to enable real time optical communication with the living brain.  The endeavor was initiated with a $2k seed grant to the California Institute for Quantitative Biosciences (QB3) and grew into a $20K showcase for the collaboration between neuroscientists, electrical engineers, and computer scientists, highlighting U.C. Berkeley's position as a preeminent leader in brain research. The conference, which already has a 50 person waitlist, will host 11 fully funded speakers including EE Associate Prof. Laura Waller,  present 25 posters, and is supported by a dozen corporate sponsors .

Armando Solar-Lezama: Academic success despite an inauspicious start

Alumnus and Mexican immigrant Armando Solar-Lezama (CS Ph.D. '08) is the subject of an MIT News article describing some of the academic obstacles he had to overcome on his path to becoming a tenured professor at MIT.  Armando's creative  approaches to his class assignments were discouraged in Mexico and despite self-educating to narrow the gap, he experienced systematic repression in high school when he moved to Texas with his family in 1997.  After he graduated from Texas A&M, he was welcomed into the Berkeley EECS graduate program.  Under the mentorship of Prof. Ras Bodik, Armando discovered the nascent area of "program synethesis," which has since blossomed into a popular field of research.  Read about Armando's challenging and inspiring journey.

Meet the most nimble-fingered robot yet

Many researchers are working on ways for robots to learn to grasp and manipulate things by practicing over and over, but the process is very time-consuming. The research work on robotic deep learning by Prof. Ken Goldberg is featured on the cover of MIT Review in an article titled "Meet the Most Nimble-Fingered Robot Yet".  Instead of practicing in the real world, Prof. Ken Goldberg and colleagues have developed a robot that learns by feeding on a data set of more than a thousand objects that includes their 3-D shape, visual appearance, and the physics of grasping them. This data set was used to train the robot’s deep-learning system. Advances in control algorithms and machine-learning approaches, together with new hardware, are steadily building a foundation on which a new generation of robots will operate.

David Culler named Interim Dean for the Division of Data Sciences

Prof. David Culler has been appointed the Interim Dean for the newly created Division of Data Sciences.  The purpose of the new division is to bring techniques to bear in statistics, mathematics, and computer science on new sources of data.  One of their goals is the creation of an undergraduate data science major and data science minor.   Prof. Culler's duties will include fostering a cooperative atmosphere among the relevant faculties; working with the administration to form an advisory board with representation of key external constituencies; advancing fundraising efforts in concert with broader campus fundraising objectives; and enlisting a team of Berkeley faculty members who will work with him to develop the initiative. He will begin his new role on July 1, 2017 for a two-year term.

Two sophomores are using AI to fight fake news on Facebook

EECS sophomore Rohan Phadte and Interdisciplinary Studies major Ash Bhat have built a Messenger bot called NewsBot to help users discern whether articles are "fake news" on Facebook. Besides determining the validity of an article, it also offers a barometer that shows where an article might fit on the left-right bias spectrum, and is one of the only tools of its kind.  The idea for the algorithm first came to them during machine-learning classes.  Although it still has some bugs, they are making updates every day and the tool will improve as more users provide feedback.   "We want people to read more than just the headline. We want them to understand what the news they see says and who it's coming from." Bhat says.  Read the article on Mic.

Stuart Russell TED talk: 3 principles for creating safer AI

CS Prof. Stuart Russell gave an engaging TED talk in April describing some of the problems in, and possible solutions for, creating a species that is smarter than humans.  He argues that building provably altruistic,  humble, and humanitarian machines might help us avoid some of the pitfalls inevitable in a future with superintelligent AI.