* * * *
Please note this event will be on Tuesday, rather than the regular
seminar day, Thursday!
* * * *

Title: Near-Optimal Sensor Placements in Gaussian Processes
Speaker: Carlos Guestrin
Location: 405 Soda Hall
Time: Thursday, Nov 2, from 11:30am to 1:00pm.

* * * *
Meeting w/ the speaker: Email ameli@cs, specifying your
availability on Tuesday
* * * *

Near-optimal Sensor Placements:
Maximizing Information while Minimizing Communication Cost


      When monitoring spatial phenomena with wireless sensor networks,
      selecting the best sensor placements is a fundamental task.
      Not only should the sensors be informative, but they
      should also be able to communicate efficiently. In this
      talk, we present a data-driven approach that addresses the
      three central aspects of this problem: measuring the
      predictive quality of a set of sensor locations, even if no
      sensors were ever placed at these locations, predicting the
      communication cost involved with these placements, and
      designing an algorithm with provable quality guarantees that
      optimizes the NP-hard tradeoff. Specifically, we use data
      from a pilot deployment to build non-parametric
      probabilistic models called Gaussian Processes (GPs)
      both for the spatial phenomena of interest and for the
      spatial variability of link qualities, which allows us to
      estimate predictive power and communication cost of unsensed
      locations. Surprisingly, uncertainty in the representation
      of link qualities plays an important role in estimating
      communication costs. Using these models, we present a novel,
      polynomial-time, data-driven algorithm, pSPIEL, which
      selects Sensor Placements at Informative and cost-Effective
      Locations. Our approach exploits two important properties of
      this problem: submodularity, formalizing the
      intuition that adding a node to a small deployment can help
      more than adding a node to a large deployment; and
      locality, where nodes that are far from each other
      provide almost independent information. Exploiting
      these properties, we prove strong approximation guarantees
      for our pSPIEL approach. We also provide extensive
      experimental validation of this practical approach on
      several real-world datasets, demonstrating significant
      advantages over existing methods.

This talk includes joint work with Andreas Krause, Ajit Singh, Anupam
Gupta, and Jon Kleinberg

Bio:
Carlos Guestrin is an assistant professor in the Center for
Automated Learning and Discovery and in the Computer Science
Department at Carnegie Mellon University. Previously, he was a
senior researcher at the Intel Research Lab in Berkeley. Carlos
Guestrin received his MSc and PhD in Computer Science from Stanford
University in 2000 and 2003, respectively, and a Mechatronics
Engineer degree from the Polytechnic School of the University of
São Paulo, Brazil, in 1998. His current research spans the areas
of planning, reasoning and learning in uncertain dynamic environments,
focusing on applications in sensor networks.
Carlos Guestrin received best paper
awards at the Information Processing in Sensor Networks (IPSN-2005),
Very Large Data Bases (VLDB-2004), and Neural Information Processing
Systems (NIPS-2003) conferences, and runner-up best paper awards at
the Uncertainty in Artificial Intelligence (UAI-2005) and Machine
Learning (ICML-2005) conferences. He is also a recipient of
the Siebel Scholarship, the Stanford Centennial Teaching
Assistant Award, the Stanford School of Engineering Fellowship, and the
FAPESP and CAPES Research Initiation Fellowships