Filters








15,404 Hits in 2.8 sec

High-dimensional Penalty Selection via Minimum Description Length Principle [article]

Kohei Miyaguchi, Kenji Yamanishi
2018 arXiv   pre-print
view of the minimum description length principle, and guides us to a good penalty function through the high-dimensional space.  ...  We tackle the problem of penalty selection of regularization on the basis of the minimum description length (MDL) principle.  ...  In this paper, we propose a novel penalty selection method that utilizes information about the objective criterion efficiently on the basis of the minimum description length (MDL) principle [Rissanen(  ... 
arXiv:1804.09904v1 fatcat:j55lamgznbe77jyxsnwfvzysaa

Minimum description length principle for linear mixed effects models

Li Li, Fang Yao, Radu V. Craiu
2014 Statistica sinica  
Abstracts The minimum description length (MDL) principle originated from data compression literature and has been considered for deriving statistical model selection procedures.  ...  We propose a class of MDL procedures that incorporate the dependence structure within individual or cluster and use data-adaptive penalties that suit both finite and infinite dimensional data generating  ...  This motivates our proposed model selection criterion for models with correlated data built upon the minimum description length (MDL) principle, as it attempts to find a good balance between AIC and  ... 
doi:10.5705/ss.2011.230 fatcat:kjbjhcp7wnanzph56cg72ydi6i

Local asymptotic coding and the minimum description length

D.P. Foster, R.A. Stine
1999 IEEE Transactions on Information Theory  
Common approximations for the minimum description length (MDL) criterion imply that the cost of adding a parameter to a model fit to n observations is about (1/2) log n bits.  ...  Abstract Common approximations for the minimum description length (MDL) criterion imply that the cost of adding a parameter to a model fit to n observations is about (1/2) log n bits.  ...  In order to select the best parametric model from among several of possibly varying dimension k, Rissanen proposes that one choose the model which obtains the minimum description length (MDL).  ... 
doi:10.1109/18.761287 fatcat:7ndhb3yslnetdhbr4gsjhmwcpy

Research on Performance Degradation Assessment Method of Train Rolling Bearings under incomplete data

Xuejun Zhao, Yong Qin, Dandan Wang, Zhipeng Wang, Limin Jia
2016 Proceedings of the 22nd International Conference on Distributed Multimedia Systems  
Secondly, the high-dimensionality of features is reduced by the principal component analysis (PCA).  ...  And then, on the basis of choosing the kernel parameter and penalty weight, a degradation method based on SVDD is proposed.  ...  Principle of support vector data description Support vector data description was originally proposed by Tax and Duin [17] .  ... 
doi:10.18293/dms2016-046 dblp:conf/dms/QinWWJZ16 fatcat:nvty6nbv5japlj7bkhb6rgpbbm

Penalized estimation in high-dimensional hidden Markov models with state-specific graphical models

Nicolas Städler, Sach Mukherjee
2013 Annals of Applied Statistics  
The methodology is adaptive and very general, applying in particular to both low- and high-dimensional settings without requiring hand tuning.  ...  Penalization is nontrivial in this latent variable setting; we propose a penalty that automatically adapts to the number of states K and the state-specific sample sizes and can cope with scaling issues  ...  penalty parameter based on the minimum description length principle].  ... 
doi:10.1214/13-aoas662 fatcat:umayk7hr6ng7ddakxs4t6f2t3m

Model Selection and the Principle of Minimum Description Length

Mark H Hansen, Bin Yu
2001 Journal of the American Statistical Association  
This paper reviews the principle of Minimum Description Length MDL for problems of model selection.  ...  In writing this review, we tried to make the descriptive philosophy of MDL natural to a statistics audience by examining classical problems in model selection.  ...  This implies that these schemes produce valid description lengths, each yielding a usable model selection criterion via the MDL principle.  ... 
doi:10.1198/016214501753168398 fatcat:a32xsezxhnexvmodusu32yue7e

An information geometry approach to shape density Minimum Description Length model selection

Adrian M. Peter, Anand Rangarajan
2011 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops)  
The free parameters associated with these estimators can then be rigorously selected using the Minimum Description Length (MDL) criterion for model selection.  ...  We consider a principled and geometric approach to selecting the model order for a class of shape density models where the square-root of the distribution is expanded in an orthogonal series.  ...  For the first time, we present a principled and geometric approach to selecting the model order of all orthogonal series estimators using the Minimum Description Length (MDL) criterion.  ... 
doi:10.1109/iccvw.2011.6130419 dblp:conf/iccvw/PeterR11 fatcat:hxb7anqxpneotgwcsueex3p65a

Robust Speaker Adaptation by Weighted Model Averaging Based on the Minimum Description Length Criterion

Xiaodong Cui, Abeer Alwan
2007 IEEE Transactions on Audio, Speech, and Language Processing  
The minimum description length (MDL) principle is applied to account for the compromise between transformation granularity and descriptive ability regarding the tying patterns of structured transformations  ...  In this paper, a robust MLLR speaker adaptation approach via weighted model averaging is investigated.  ...  Hansen for discussions on model selection.  ... 
doi:10.1109/tasl.2006.876773 fatcat:a7rwj6hyzzg6jmn6cui6ltsn6a

Detecting abrupt changes in the spectra of high-energy astrophysical sources

Raymond K. W. Wong, Vinay L. Kashyap, Thomas C. M. Lee, David A. van Dyk
2016 Annals of Applied Statistics  
Because the model is over parameterized we employ an 1 penalty. The tuning parameter in the penalty and the number of change points are determined via the minimum description length principle.  ...  Our method is validated via a series of simulation studies and its practical utility is illustrated in the analysis of the ultra-fast rotating yellow giant star known as FK Com.  ...  Because the model is over parameterized we employ an 1 penalty. The tuning parameter in the penalty and the number of change points are determined via the minimum description length principle.  ... 
doi:10.1214/16-aoas933 fatcat:hrlb7dyw4jde5ibdeje65hb4ni

Explaining Anomalies in Groups with Characterizing Subspace Rules [article]

Meghanath Macha, Leman Akoglu
2018 arXiv   pre-print
Namely, it can unearth anomalous patterns (i) of multiple different types, (ii) hidden in arbitrary subspaces of a high dimensional space, (iii) interpretable by the analysts, (iv) different from normal  ...  We consider a complementary problem that has a much sparser literature: anomaly description.  ...  To decide which patterns describe the anomalies most succinctly, we introduce an encoding scheme based on the Minimum Description Length (MDL) principle [52] .  ... 
arXiv:1708.05929v4 fatcat:v5vufosjurdyjfuipt5df5jg7i

Are Bitcoins price predictable? Evidence from machine learning techniques using technical indicators [article]

Samuel Asante Gyamerah
2019 arXiv   pre-print
The prediction models employed key and high dimensional technical indicators as the predictors.  ...  These values show a high degree of reliability in predicting the price of Bitcoin using the stacking ensemble model.  ...  The elastic-net penalty combines the ridge and lasso penalty; if features are correlated in groups, a γ = 0.5 is likely to select the groups in or out simultaneously.  ... 
arXiv:1909.01268v1 fatcat:z7ctx7brlngahb4ras45cvewnu

An information-theoretic framework for semantic-multimedia retrieval

João Magalhães, Stefan Rüger
2010 ACM Transactions on Information Systems  
a vocabulary size we use the minimum description length principle to select its optimal size.  ...  All results were obtained with a 972 dimensional multi-modal feature space selected by the minimum description length criterion.  ... 
doi:10.1145/1852102.1852105 fatcat:od72dimsonegxdy3hy3aoj4inq

BeatLex: Summarizing and Forecasting Time Series with Patterns [chapter]

Bryan Hooi, Shenghua Liu, Asim Smailagic, Christos Faloutsos
2017 Lecture Notes in Computer Science  
and parameterfree, as it is based on the Minimum Description Length principle of summarizing the data by compressing it using as few bits as possible, and automatically tunes all its parameters; 4) general  ...  fast and online, requiring linear time in the data size and bounded memory; 2) effective, outperforming competing algorithms in labelling accuracy by 5.3 times, and forecasting accuracy by 1.8 times; 3) principled  ...  -Principled and parameter-free: BEATLEX is fit using the Minimum Description Length (MDL) principle, and automatically tunes its parameters.  ... 
doi:10.1007/978-3-319-71246-8_1 fatcat:m32rzibgtbatnd3i7sxpprn32a

Simultaneous Noise Suppression and Signal Compression Using a Library of Orthonormal Bases and the Minimum Description Length Criterion [chapter]

Naoki Saito
1994 Wavelet Analysis and Its Applications  
description length (MDL).  ...  Description Length (MDL) criterion for discriminating signal from noise.  ...  THE MINIMUM DESCRIPTION LENGTH PRINCIPLE To satisfy the above mentioned conflicting demands, we need a model selection criterion.  ... 
doi:10.1016/b978-0-08-052087-2.50017-7 fatcat:3gx7gdypvzaevewccj2ri2zlhu

Simultaneous noise suppression and signal compression using a library of orthonormal bases and the minimum-description-length criterion

Naoki Saito, Harold H. Szu
1994 Wavelet Applications  
description length (MDL).  ...  Description Length (MDL) criterion for discriminating signal from noise.  ...  THE MINIMUM DESCRIPTION LENGTH PRINCIPLE To satisfy the above mentioned conflicting demands, we need a model selection criterion.  ... 
doi:10.1117/12.170027 fatcat:au6m5z256fdezgpnv6ga3zgk5a
« Previous Showing results 1 — 15 out of 15,404 results