• Publications
  • Influence
A Probabilistic Approach to WLAN User Location Estimation
TLDR
We estimate the location of a WLAN user based on radio signal strength measurements performed by the user's mobile terminal based on a probabilistic framework for solving the location estimation problem. Expand
A Statistical Modeling Approach to Location Estimation
TLDR
We present an approach to location estimation based on a statistical signal power model that is different from the prevailing geometric one. Expand
Conditional NML Universal Models
  • J. Rissanen, T. Roos
  • Mathematics
  • Information Theory and Applications Workshop
  • 22 October 2007
The NML (normalized maximum likelihood) universal model has certain minmax optimal properties but it has two shortcomings: the normalizing coefficient can be evaluated in a closed form only forExpand
Evaluating methods for computer-assisted stemmatology using artificial benchmark data sets
TLDR
We describe an experiment involving three artificial benchmark data sets to which a number of computer-assisted stemmatology methods were applied. Expand
Learning locally minimax optimal Bayesian networks
TLDR
We consider the problem of learning Bayesian network models in a non-informative setting, where the only available information is a set of observational data, and no background knowledge is available. Expand
Factorized normalized maximum likelihood criterion for learning Bayesian network structures
TLDR
We introduce a new probabilistic scoring criterion, the factorized normalized maximum likelihood, for learning Bayesian network structures from complete discrete data. Expand
On Discriminative Bayesian Network Classifiers and Logistic Regression
TLDR
We show that if the network structure satisfies a certain graph-theoretic condition, the corresponding conditional likelihood maximization problem is equivalent to logistic regression based on certain statistics of the data—different network structures leading to different statistics. Expand
MDL Denoising Revisited
TLDR
We refine and extend an earlier minimum description length (MDL) denoising criterion for wavelet-based denoises and introduce soft thresholding inspired by predictive universal coding with weighted mixtures. Expand
Discriminative Learning of Bayesian Networks via Factorized Conditional Log-Likelihood
TLDR
We propose an efficient and parameter-free scoring criterion, the factorized conditional log-likelihood (fCLL), for learning Bayesian network classifiers. Expand
On Sequentially Normalized Maximum Likelihood Models
The important normalized maximum likelihood (NML) distribution is obtained via a normalization over all sequences of given length. It has two short-comings: the resulting model is usually not aExpand
...
1
2
3
4
5
...