Welcome to the Select Lab

Our main long-term research goal is developing efficient algorithms and methods for designing, analyzing, and controlling complex real-world systems. To achieve this goal, our research spans the entire spectrum from theoretical foundations to real-world applications.

Algorithmically, we focus on machine learning problems, where a system must sense the world, learn from experience and make effective long-term decisions. The techniques we develop encompass a wide range of topics, including probabilistic graphical models (e.g., Bayesian and Markov networks), active learning and value of information, distributed algorithms, probabilistic inference, decision making under uncertainty (e.g., partially observable Markov decision problems), online learning and game theory.

Our main applications are sensor networks, computer systems and robotics. Our current projects include real world deployments of sensors to monitor pollution in lakes and rivers, detect pathogens in our water distribution systems, build technology to help our aging and disabled population, and perform diagnosis in complex computer systems and networks.

Recent Events

2008

Tutorial: Beyond Convexity - Submodularity in Machine Learning

Andreas Krause and Carlos Guestrin will present a tutorial at ICML 2008 in Helsinki.
2008

ONR Young Investigator Award

Carlos Guestrin was granted an ONR Young Investigator Award.
1/28/2008

From ranking blogs to predicting posture

Here is a recent MSNBC article on the CASCADES and sensor placement projects.
12/14/2007

Sajid Siddiqi Thesis Proposal

Sajid Siddiqi will present his thesis proposal on "Latent Variable and Predictive Models of Dynamical Systems."
11/27/2007

Honorable Mention for Outstanding Student Paper Award

Jonathan Huang, Carlos Guestrin and Leonidas Guibas received Honorable Mention for the Outstanding Student Paper Award at NIPS 2007 for the paper "Efficient Inference for Distributions on Permutations"
11/26/2007

Picking which blogs to read

Which blogs should we read to be most up to date on the Blogosphere? Where should we place sensors to detect contaminations in water networks? The CASCADES project attempts to answer these questions in a unified way. Here are some discussions on the project. Also see the news release and this post.
10/25/2007

Andreas Krause Thesis Proposal

Andreas Krause presented his thesis proposal on "Nonmyopic Observation Selection: Theory and Applications".
2007

NSF Fellowship Awards

Joseph Gonzalez and Joseph Bradely were both awarded NSF Fellowships graduate research fellowships.
07/2007

2007 IJCAII-JAIR Best Paper prize

Carlos Guestrin, Daphne Koller, Ron Parr and Shobha Venkataraman received the 2007 IJCAII-JAIR Best Paper prize in the Journal of Artificial Intelligence Research (JAIR) for "Efficient Solution Algorithms for Factored MDPs." This prize is awarded to an outstanding paper published in JAIR in the preceding five calendar years.
2007

IBM Faculty Fellowship

Carlos Guestrin was awarded a 2007 IBM Faculty Fellowship.
2007

AT&T Fellowship Award

Joseph Gonzalez was awarded the AT&T Labs research Fellowship.
08/2007

Best Student Paper Award

Jure Leskovec, Andreas Krause, Carlos Guestrin, Christos Faloutsos, Jeanne VanBriesen and Natalie Glance were awarded the Best Student Paper Award at ACM SIGKDD 2007 for the paper "Cost-effective Outbreak Detection in Networks".
01/2007

Microsoft Research Fellowship Award

Andreas Krause was awarded the Microsoft Research Graduate Fellowship.
2007

NSF Career Award

Carlos Guestrin was granted an NSF Career Award.