HomeProjects
Projects
Computational Sciences and Engineering
A great many fields of science, engineering, finance and social science are embracing modeling, simulation, and data analysis as necessary tools to advance their fields. Sometimes this is driven by the march of Moore’s Law providing computational power that makes simulations possible that were not possible before; it is also driven by the availability of large data sets not available before that require extensive computation to understand.
Socio-Political
Our project aims at providing the general public with easy-to-interpret statistical analyses of large-scale data sets pertaining to political life, with a focus on voting records, online news sources, and polling data. Our approach is to marry sparse (interpretable) statistical inference principles with advanced distributed computational and optimization techniques.
The Econometrics Laboratory (EML)
The Econometrics Laboratory (EML) is the computing and statistical resource for the Economics Department’s faculty and graduate students. EML’s founding Director is Professor Daniel McFadden, the 2000 Nobel Laureate in Economics. The creation of the EML was motivated by Professor McFadden’s vision and it is dedicated to education and research in the field of computationally intensive econometrics, utilizing and advancing state-ofthe- art econometric methods, software, and hardware. The EML anticipates that the access to the computing power of the Yahoo!
Climate
Regional climate simulations of historic and projected climate at 30-km for the western U.S. and nested to 10-km California/Nevada. This work is part of the California Assessment through CEC funding in response and as a contribution to AB32. We are performing 10 year integrations for several sets of forcing conditions based on IPCC AR4 scenarios and output.
Neuroscience
The project goal is to determine how local populations of neurons in visual cortex process dynamic natural stimuli. Research will progress along two principal directions.
Computational Characterization of Nanoporous Materials – Zeolites
Porous materials have many important applications in the chemical industry. Zeolites, for example, are used for gas separation as well as cracking catalysts in oil refining, alkylation and isomerization reactions catalysts. Although the number of possible zeolite structures has been estimated to be more than 2.5 million, only about 180 structures have been synthesized to date. Our long-term goal is to develop methods that will allow the identification of structures with optimal performance for specific applications, with a focus on predicting new materials for CO2 separati
Phylogenomic tools to predict protein 3D structure and functional sites
The fundamental aim of this project is to develop a community resource providing a pipeline including powerful methods for protein structure prediction, protein superfamily phylogeny reconstruction and identification of functionally important positions. The resource will provide statistical models tuned for specific types of functional residues (e.g., catalytic residues, ligand-binding positions, allosteric sites, and metal-binding residues).
The Berkeley Transients Classification Pipeline (TCP)
The Berkeley Transients Classification Pipeline (TCP) is an astronomical time series classification project which identifies flux varying stellar sources in real-time data streams. Upon identification of scientifically interesting sources, the pipeline emits source information to robotic telescopes for immediate and automated follow-up using complimentary filters and instruments. Since the TCP incorporates several historical survey data sets, as well nightly data streams, the volume of data accessed during classifier training an
Energy Efficient Network Building Control
Project Summary: This project focuses on the modeling and the control of the thermal energy storage on the campus of the University of California, Merced, USA. The campus has been designed to be a ''living laboratory'' and has a significantly enhanced level of instrumentation in order to support the development and demonstration of energy-efficient technologies and practices.
Cloud Computing for Perception-Based Intelligent Decision Systems
Project Summary: Everyday, we are creating more and more data. Most of today's Scientific applications create TeraB data every year and some even creates over 10s of TeraB and this will increase in coming years and will reach PetaB data. Mining these information and creating Decision Tree will be an important area of research.
Multiscale Simulations of Soft Contact and Adhesion of Stem cells
Project Summary: The research project is about using multi-scale computational methods to simulate soft contact and adhesion of stem cells with various substrates in order to discover the mysterious mechano-transduction mechanism of stem cells, which is the key governing factor of the stem cell differentiation and lineage. The project has involved with intensive 2D and 3D large scale simulations of various coarse-grained cell models of multiple components. Shaofan Li, University of California, Berkeley
Statistical Machine Learning
Project Summary: Prof. Jordan's group works in a number of fundamental topics in machine learning, including graphical modeling, clustering, classification, dimension reduction, spectral analysis, independent component analysis, reinforcement learning and link analysis.The group also works on applications of machine learning, in areas such as distributed computer systems, natural language processing, signal processing, genomics, proteomics, robot control, information retrieval, image analysis, analog circuit design and software engineering.
Bringing the power of High-End Computing to Simultaneous Multiple Sequence Alignment and Phylogenetic Tree Construction
The fundamental aim of the work outlined in this proposal is to create a high-end computing solution to large-scale estimation of multiple sequence alignments and phylogenetic trees for protein superfamilies (families of proteins related by gene duplication, and thus containing groups of paralogous proteins).
Creating Decision Tree for TeraB and PetaB Socio-Economic-Policy and Spatio-Temporal Massive Scientific Data
Every day, we are creating more and more data. Most of today's
Socio-Economic-Policy and Scientific applications/ create TeraB data
every year and some even creates over 10s of TeraB and this will
increase in coming years and will reach PetaB data. Mining these
information and creating evolutionary computing-based decision tree from
imprecise-perception-based /Socio-Economic-Policy-Scientific/ will be an
important area of research.