AMPLab researchers Sanjay Krishnan, Prof. Michael Franklin, Prof. Ken Goldberg, Eugene Wu, and Jiannan Wang have developed ActiveClean, a system that uses machine learning to improve the process of removing dirty data by analyzing a user's prediction model to decide which mistakes to edit first, while updating the model as it works.  The demonstration paper titled "ActiveClean: An Interactive Data Cleaning Framework For Modern Machine Learning" received the Best Demo Award at SIGMOD 2016.

ActiveClean is profiled in an I Programmer article and the development team led byEugene Wu (now at Columbia) will present its research on Sept. 7 in New Delhi, at the 2016 conference on Very Large Data Bases.

Profs. Sanjit Seshia, Ruzena Bajcsy, Shankar Sastry, Björn Hartmann, Claire Tomlin and Tom Griffiths are the principal investigators of a new large National Science Foundation project that will tackle the problem of designing “human Cyber-Physical Systems (h-CPS)”, cyber-physical systems that work in concert with humans. The research outcome of the project, called Verified Human Interfaces, Control, and Learning for Semi-Autonomous Systems, or VeHICaL, will have applications in emerging technologies such as semi-autonomous cars and autonomous aerial vehicles (drones). NSF has awarded $4.6M for this project.

Profs. Shankar Sastry and Pravin Varaiya have been elected as International Federation of Automatic Control (IFAC) Fellows. The IFAC Fellow Award is given to persons who have made outstanding and extraordinary contributions in the field of interest of IFAC, in the role as an Engineer/Scientist. Prof. Sastry is recognized for contributions to research and education in the areas of robotic manipulation, adaptive control, nonlinear feedback, and hybrid systems, and Prof. Varaiya is recognized for contributions to control theory, with applications to transportation, power, communications and economic systems.

The research of postdoctoral scholar Oscar Beijbom (supervised by Prof. Trevor Darrell at the Berkeley Vision and Learning Center) is being used to help coral researchers speed up the under-sea mapping of reefs around the world.  Beijbom's neural-networking algorithm is part of a software project,  featured in a Nature article, that can zip through and classify over a million photographs in a matter of months.

The Center for Human-Compatible Artificial Intelligence, which will focus on ensuring that AI systems are beneficial to humans, is being lead by  Prof. Stuart Russell, a long-time advocate of incorporating human values in the design of AI.  Associate Prof. Pieter Abbeel and Assistant Prof. Anca Dragan will serve as co-principal investigators along with cognitive scientist Tom Griffiths and faculty from Cornell and the University of Michigan.  The center was made possible by a grant of $5.5 million from the Open Philanthropy Project, as well as grants from the Leverhulme Trust and the Future of Life Institute.

Prof. Sergey Levine, EECS alumni Oriol Vinyals and EECS postdoc Wei Gao (working with Ali Javey) have been named on MIT Technology Review’s 2016 TR35 (Innovators Under 35) who push the edge of science, creating new approaches to tackling technology challenges. In the “Pioneers” category Prof. Levine teaches robots to watch and learn from their own successes, supervising it’s own learning, and Oriol Vinyals is working to create computers that can teach themselves how to play and win complex games by enabling them to learn from experience. In the “Inventors” category, Wei Gao is building sweatbands that monitor your health on a molecular level.

Prof. Sylvia Ratnasamy has been selected by Networking Women for their inaugural list of “10 women in networking/communications that you should WATCH”. Over 100 people around the world submitted nominations for this list and the women nominated have all had impact on the networking field, early in their careers. Prof. Ratnasamy’s research focuses on the design and implementation of networked systems. She co-lead the SPAN Center for networking research. She is a recipient of the ACM Grace Murray Hopper award, the ACM SIGCOMM Test-of-Time award, the ACM SIGCOMM Rising Star award, and the Alfred P. Sloan research fellowship.

Prof. Marti Hearst and a team in the I School are featured in a Berkeley School of Information article titled “Teaching a Computer to Write Poetry”. The discipline of Natural Language Processing aims to analyze and understand human language.  Computers are learning to respond using human language, but poetry presents unique challenges. Poetry evokes feelings and emotion more than just the written word, and sonnets follow a strict scheme of rhyme, meter and metaphor.  Computer-generated sonnets from this team placed second in the 2016 “PoetiX”, a competition of only computer-generated traditional sonnets: fourteen line poems, in iambic pentameter, in either “Shakespearean” or “Petrarchan” form.

Our Electrical Engineering and Computer Science Divisions both ranked third in the 2016 Academic Ranking of World Universities.  UC Berkeley was once again named the top public university worldwide.  The ARWU annually ranks more than 1,200 universities on the quality of both faculty and research.

EECS came in third after Stanford and MIT in both fields.  The CS top five was rounded out by Harvard and Princeton,  while Nanyang University and Georgia Tech followed Berkeley in the EE top five.

Incoming CS Assistant Professor Jonathan Ragan-Kelley, alumnus Shoaib Kamil (Ph.D. CS 2012 under Profs. Armando Fox and Kathy Yelick) and alumnus Wojciech Matusik (B.S. EECS 1997), along with other researchers at MIT CSAIL, Adobe, U. of Toronto, Texas A&M, and U. of Texas have developed Simit,  a programming language that can speed up computer simulations 200-fold or reduce the code they require by 90 percent.

The language has applications outside simulations, and there are even plans for it to augment machine learning, data analytics, optimization and robotics in addition to a version of Google's PageRank algorithm.