A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf
.
Filters
Estimating divergence functionals and the likelihood ratio by penalized convex risk minimization
2007
Neural Information Processing Systems
Our method is based on a variational characterization of f -divergences, which turns the estimation into a penalized convex risk minimization problem. ...
We develop and analyze an algorithm for nonparametric estimation of divergence functionals and the density ratio of two probability distributions. ...
In this paper, we estimate the likelihood ratio and the KL divergence by optimizing a penalized convex risk. ...
dblp:conf/nips/NguyenWJ07
fatcat:hoxwvye2ije33czieq4xmoildi
Convex Multiple-Instance Learning by Estimating Likelihood Ratio
2010
Neural Information Processing Systems
Theoretically, we prove a quantitative relationship between the risk estimated under the 0-1 classification loss, and under a loss function for likelihood ratio. ...
This is casted as joint estimation of both a likelihood ratio predictor and the target (likelihood ratio variable) for instances. ...
Acknowledgements This work is supported, in part, by the European Commission, under a Marie Curie Excellence Grant MCEXT-025481. ...
dblp:conf/nips/LiS10
fatcat:5izppstt7nhhfc2ra7acgm4sku
New Developments in Statistical Information Theory Based on Entropy and Divergence Measures
2019
Entropy
In the last decades the interest in statistical methods based on information measures and particularly in pseudodistances or divergences has grown substantially [...] ...
Since no optimality holds for the aggregation of likelihood ratio tests, a similar procedure is proposed, replacing the individual likelihood ratio by some divergence based test statistics. ...
Besides the theoretical results, they have constructed an efficient algorithm, in which we minimize a convex loss function at each iteration. ...
doi:10.3390/e21040391
pmid:33267105
fatcat:jc37cuc4yjamhhdlddrppmkzuu
Penalized Bregman divergence for large-dimensional regression and classification
2010
Biometrika
We introduce the penalized Bregman divergence by replacing the negative loglikelihood in the conventional penalized likelihood with Bregman divergence, which encompasses many commonly used loss functions ...
It is shown that the resulting penalized estimator, combined with appropriate penalties, achieves the same oracle property as the penalized likelihood estimator, but asymptotically does not rely on the ...
In that case, if convex penalties are used in (6), then n ( β) is necessarily convex in β, and hence the local minimizer β E is the unique global penalized Bregman divergence estimator. ...
doi:10.1093/biomet/asq033
pmid:22822248
pmcid:PMC3372245
fatcat:lyvdgqpvozbkpkuygxirhmisea
Statistical models, likelihood, penalized likelihood and hierarchical likelihood
[article]
2008
arXiv
pre-print
The Kullback-Leibler divergence is referred to repeatedly, for defining the misspecification risk of a model, for grounding the likelihood and the likelihood crossvalidation which can be used for choosing ...
Families of penalized likelihood and sieves estimators are shown to be equivalent. The similarity of these likelihood with a posteriori distributions in a Bayesian approach is considered. ...
I would like to thank Anne Gégout-Petit for helpful comments on the manuscript. ...
arXiv:0808.4042v1
fatcat:zsxiwq2nhrdbplybv3jos3aqqa
Statistical models: Conventional, penalized and hierarchical likelihood
2009
Statistics Survey
The Kullback-Leibler divergence is referred to repeatedly in the literature, for defining the misspecification risk of a model and for grounding the likelihood and the likelihood cross-validation, which ...
Families of penalized likelihood and particular sieves estimators are shown to be equivalent. The similarity of these likelihoods with a posteriori distributions in a Bayesian approach is considered. ...
Acknowledgements I would like to thank Anne Gégout-Petit for helpful comments on the manuscript.
D. Commenges/Statistical models and likelihoods ...
doi:10.1214/08-ss039
fatcat:imjm3d5dsfdqljdcm7mddjb43m
MDL, penalized likelihood, and statistical risk
2008
2008 IEEE Information Theory Workshop
Penalized likelihood risk bounds should capture the tradeoff of Kullback-Leibler approximation error and penalty. ...
This is examined for 1 penalized least squares in the manuscripts [80] and [39] and for 1 penalized likelihood in the present paper. ...
RISK AND RESOLVABILITY FOR COUNTABLEF Here we recall risk bounds for penalized likelihood with a countableF. ...
doi:10.1109/itw.2008.4578660
dblp:conf/itw/BarronHLL08
fatcat:zvxpvgjpgjejjjv32ywoxlu3u4
Penalized high-dimensional empirical likelihood
2010
Biometrika
By using an appropriate penalty function, we show that penalized empirical likelihood has the oracle property. ...
We propose penalized empirical likelihood for parameter estimation and variable selection for problems with diverging numbers of parameters. ...
For (1) and (3) to have solutions, μ needs to be in the convex hull formed by {X i } n i=1 . Therefore, the penalized empirical likelihood estimatorμ must lie within the convex hull of the data. ...
doi:10.1093/biomet/asq057
fatcat:s5faeofjrfgzdmbqud62j6by7e
A Practical Transfer Learning Algorithm for Face Verification
2013
2013 IEEE International Conference on Computer Vision
Based upon a surprisingly simple generative Bayesian model, our approach combines a KL-divergencebased regularizer/prior with a robust likelihood function leading to a scalable implementation via the EM ...
As justification for our design choices, we later use principles from convex analysis to recast our algorithm as an equivalent structured rank minimization problem leading to a number of interesting insights ...
We may now optimize the objective function in (3) by iteratively computing the expectation of the latent variables (E step) and updating the parameters by maximizing the expected penalized log-likelihood ...
doi:10.1109/iccv.2013.398
dblp:conf/iccv/CaoWWD013
fatcat:6w46kxokljahrpvde6air6zqzq
A Selective Overview of Variable Selection in High Dimensional Feature Space
2010
Statistica sinica
The properties of non-concave penalized likelihood and its roles in high dimensional statistical modeling are emphasized. ...
What limits of the dimensionality such methods can handle, what the role of penalty functions is, and what the statistical properties are rapidly drive the advances of the field. ...
Acknowledgments Fan's research was partially supported by NSF Grants DMS-0704337 and DMS-0714554 and NIH Grant R01-GM072611. ...
pmid:21572976
pmcid:PMC3092303
fatcat:kf3kbxcyozdaleej4i63y63phm
Wald-Kernel: Learning to Aggregate Information for Sequential Inference
[article]
2017
arXiv
pre-print
We formulate the problem as a constrained likelihood ratio estimation which can be solved efficiently through convex optimization by imposing Reproducing Kernel Hilbert Space (RKHS) structure on the log-likelihood ...
The proposed algorithm, namely Wald-Kernel, is tested on a synthetic data set and two real world data sets, together with previous approaches for likelihood ratio estimation. ...
[7] derived variational characterizations of f -divergences which enabled estimation of divergence functionals and likelihood ratios through convex risk minimization. ...
arXiv:1508.07964v3
fatcat:vrcl5rwg25ga3aiq5zqo5ltuk4
Collaborative likelihood-ratio estimation over graphs
[article]
2024
arXiv
pre-print
Assuming we have iid observations from two unknown probability density functions (pdfs), p and q, the likelihood-ratio estimation (LRE) is an elegant approach to compare the two pdfs only by relying on ...
from two unknown node-specific pdfs, p_v and q_v, and the goal is to estimate for each node the likelihood-ratio between both pdfs by also taking into account the information provided by the graph structure ...
Acknowledgments The authors acknowledge support from the Industrial Data Analytics and Machine Learning Chair hosted at ENS Paris-Saclay, Université Paris-Saclay. ...
arXiv:2205.14461v2
fatcat:p77bnoslcndkhdp5qk5gbgkal4
Relative Novelty Detection
2009
Journal of machine learning research
By design this is dependent on the underlying measure of the space. ...
In this paper we derive a formulation which is able to address this problem by allowing for a reference measure to be given in the form of a sample from an alternative distribution. ...
Variational Decomposition Divergences between distributions, say p and q which can be expressed as the expectation over a function of a likelihood ratio can be estimated directly by solving a convex minimization ...
dblp:journals/jmlr/JSmolaST09
fatcat:3akzacuufzgoflczqptavuyf4a
A Selective Overview of Variable Selection in High Dimensional Feature Space (Invited Review Article)
[article]
2009
arXiv
pre-print
The properties of non-concave penalized likelihood and its roles in high dimensional statistical modeling are emphasized. ...
What limits of the dimensionality such methods can handle, what the role of penalty functions is, and what the statistical properties are rapidly drive the advances of the field. ...
Oracle property What are the sampling properties of penalized least squares (4) and penalized likelihood estimation (2) when the penalty function p λ is no longer convex? ...
arXiv:0910.1122v1
fatcat:elzacdq7ircbjepok7bbkk5zyi
Penalized likelihood regression for generalized linear models with non-quadratic penalties
2009
Annals of the Institute of Statistical Mathematics
This is the case when formulating penalized likelihood regression for exponential families. ...
One of the popular method for fitting a regression function is regularization: minimizing an objective function which enforces a roughness penalty in addition to coherence with the data. ...
P6/03 of the Federal Science Policy, Belgium, is acknowledged. The second author also gratefully acknowledges financial support by the GOA/07/04-project of the Research Fund KU Leuven. ...
doi:10.1007/s10463-009-0242-4
fatcat:53s7eednjzfehkrtudgonavofy
« Previous
Showing results 1 — 15 out of 2,000 results