• Publications
  • Influence
Constructing free-energy approximations and generalized belief propagation algorithms
This work explains how to obtain region-based free energy approximations that improve the Bethe approximation, and corresponding generalized belief propagation (GBP) algorithms, and describes empirical results showing that GBP can significantly outperform BP.
Understanding belief propagation and its generalizations
It is shown that BP can only converge to a fixed point that is also a stationary point of the Bethe approximation to the free energy, which enables connections to be made with variational approaches to approximate inference.
Generalized Belief Propagation
It is shown that BP can only converge to a stationary point of an approximate free energy, known as the Bethe free energy in statistical physics, and generalized belief propagation (GBP) versions of these Kikuchi approximations are derived.
Performance of Fountain Codes in Collaborative Relay Networks
It is shown that the concept of mutual- information-accumulation can be realized with the help of fountain codes, and leads to a lower energy expenditure and a lower transmission time than energy accumulation.
Bethe free energy, Kikuchi approximations, and belief propagation algorithms
It is shown that BP canonly converge to a stationarypoint of an approximatefree energy, known astheBethefreeenergy in statisticalphysics, as of its release on April 4, 2001.
Iterative Quantization Using Codes On Graphs
The results suggest that graphical models may yield near optimal codes in source coding as well as in channel coding and that duality plays a key role in such constructions.
Rateless codes on noisy channels
This paper studies the performance of two classes of rateless codes (LTand Raptor codes) on noisy channels such as the BSC and the AWGNC. We find that Raptor codes outperform LT codes, and have good
Tangible interaction + graphical interpretation: a new approach to 3D modeling
It is argued that construction toys, suitably instrumented or sensed, could be the inspiration for a new generation of easy-to-use, tangible modeling systems—especially if the tangible modeling is combined with graphical-interpretation techniques for enhancing nascent models automatically.
An Idiosyncratic Journey Beyond Mean Field Theory
I try to clarify the relationships between different ways of deriving or correcting mean field theory, and present ”translations” between the language of physicists and that of computer scientists.
Feature extraction for a Slepian-Wolf biometric system using LDPC codes
An information-theoretically secure biometric storage system using graph-based error correcting codes in a Slepian-Wolf coding framework, which achieves very low false accept rates and low false reject rates on a fingerprint database.