Dissertation Talk: Exploratory model analysis for machine learning

405 Soda Hall
  • Biye Jiang, UC Berkeley, Department of EECS
Machine learning is growing in importance in many different fields. However, it is still very hard for users to tune hyper-parameters when optimizing their models, or perform a comprehensive and interpretable diagnosis for complex models like deep neural nets. Existing developer tool like TensorBoard only provides limited functionality which usually visualizes model statistics based on metrics...

Jonathan Carter Guest Lecture (4/26): Quantum Computing: The Ultimate Accelerator for HPC?

306 Soda Hall
  • Jonathan Carter, Lawrence Berkeley National Laboratory
As the last lecture of the Applications of Parallel Computers (CS267) class, Jonathan Carter will give a seminar on "Quantum Computing: The Ultimate Accelerator for HPC?". The lecture is at 306 Soda Hall, 11am-12:30pm on Thursday, April 26. Dr. Carter is the Computing Sciences Deputy of Science at the Lawrence Berkeley National Laboratory. An abstract is as follows. "Quantum Computing has...

Dissertation Talk: Evaluation and Design of Robust Neural Network Defenses

405 Soda Hall
  • Nicholas Carlini, University of California, Berkeley
Neural networks provide state-of-the-art results for most machine learning tasks. Unfortunately, neural networks used for classification are vulnerable to test-time evasion attacks (i.e., adversarial examples): given an input x and any target classification t, it is possible to find a new input x' that is similar to x but classified as t. This makes it difficult to apply neural networks in...

Dissertation talk: Classical delegation and verification of quantum computations

540AB Cory Hall
  • Urmila Mahadev
Due to recent advances in quantum computing, two related open questions have become increasingly important. First, can a classical computer delegate a quantum computation without compromising privacy? Next, is it possible for a classical computer to verify the result of a quantum computation? In this talk, we present methods allowing a classical computer to achieve both of these cryptographic...

Dissertation Talk: Learning from Language

730 Sutardja Dai Hall
  • Jacob Andreas, UC Berkeley
Natural language is built from a library of concepts and compositional operators that provide a rich source of information about how humans understand the world. Can this information help us build better machine learning models? In this talk, we'll explore three ways of integrating compositional linguistic structure and learning: using language as a source of modular reasoning operators for...

Jacobs Spring Design Showcase

Jacobs Hall
On Wednesday, May 2, and Thursday, May 3, join the Jacobs Institute for the Jacobs Spring Design Showcase. At this lively open house, you can meet student designers, check out innovations in fields from health to mobility, and celebrate the semester over conversation and refreshments.

Dissertation Talk: Program Synthesis for Systems Biology

405 Soda Hall
  • Ali Sinan Koksal, EECS, UC Berkeley
Cell signaling controls basic cellular activities and coordinates cell actions, such as cell differentiation, division and growth. Consequently, errors in cellular signaling are responsible for diseases such as cancer, autoimmunity, and diabetes. Executable biology describes mechanistic models of biological processes in a formal language that is dynamic and executable by a computer. Models in...

EECS Student Awards Ceremony

Banatao Auditorium Sutardja Dai Hall
Each year the EECS Student Awards Committee selects winners for the EECS department awards which recognize and honor the ability and character of our students.