Filters








40,296 Hits in 4.4 sec

Bayesian adaptive matrix factorization with automatic model selection

Peixian Chen, Naiyan Wang, Nevin L. Zhang, Dit-Yan Yeung
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We address these two issues simultaneously in this paper by proposing a robust non-parametric Bayesian adaptive matrix factorization (AMF) model.  ...  Nevertheless, the reliability of existing matrix factorization methods is often hard to guarantee due to challenges brought by such model selection issues as selecting the noise model and determining the  ...  In this paper, we propose our novel non-parametric full Bayesian model for adaptive matrix factorization (AMF).  ... 
doi:10.1109/cvpr.2015.7298733 dblp:conf/cvpr/ChenWZY15 fatcat:fvook2hfnbaobnompxzsawvsqe

Local Factor Analysis with Automatic Model Selection: A Comparative Study and Digits Recognition Application [chapter]

Lei Shi, Lei Xu
2006 Lecture Notes in Computer Science  
On the level of regularization, an data smoothing based regularization technique is adapted into this automatic LFA-HDS learning for problems with small sample sizes, while on the level of model selection  ...  , the proposed automatic LFA-HDS algorithm makes parameter learning with automatic determination of both the component number and the factor number in each component.  ...  compared with typical model selection criteria via the conventional two-phase procedure and the IMoFA approach.  ... 
doi:10.1007/11840930_27 fatcat:w67mmfaktfgvbauudjtxbqjkf4

Automatic Rank Determination in Projective Nonnegative Matrix Factorization [chapter]

Zhirong Yang, Zhanxing Zhu, Erkki Oja
2010 Lecture Notes in Computer Science  
Projective Nonnegative Matrix Factorization (PNMF) has demonstrated advantages in both sparse feature extraction and clustering.  ...  In this paper, we propose a method called ARDPNMF to automatically determine the column rank in PNMF. Our method is based on automatic relevance determination (ARD) with Jeffrey's prior.  ...  Model Selection in NMF In NMF, Tan and Févotte [12] addressed the model selection problem based on automatic relevance determination.  ... 
doi:10.1007/978-3-642-15995-4_64 fatcat:oehv3aach5dhhixgobdxfw72yy

Bayesian Robust Tensor Factorization for Incomplete Multiway Data

Qibin Zhao, Guoxu Zhou, Liqing Zhang, Andrzej Cichocki, Shun-Ichi Amari
2016 IEEE Transactions on Neural Networks and Learning Systems  
In contrast to existing related works, our method can perform model selection automatically and implicitly without need of tuning parameters.  ...  For model learning, we develop an efficient closed-form variational inference under a fully Bayesian treatment, which can effectively prevent the overfitting problem and scales linearly with data size.  ...  The Matlab codes for demonstrations on synthetic data and video background modeling are also provided at http://www.bsp.brain.riken.jp/ ∼ qibin/homepage/Software files/ BayesRobustTensorFactorizationP.rar  ... 
doi:10.1109/tnnls.2015.2423694 pmid:26068876 fatcat:6ur3z46hfreqjhucez7l4vriku

Hyperparameter optimization for improving recognition efficiency of an adaptive learning system

Diem-Phuc Tran, Gia-Nhu Nguyen, Van-Dung Hoang
2020 IEEE Access  
The proposed method was developed from a framework searching a set of learning hyperparameters based on the evaluation of the previous CNN model with the collected dataset during the movement of advanced  ...  The proposed solution consists of some major steps in a loop of adaptive learning system, such as (1) training an initial recognition model, (2) locating and receiving image data of different cases of  ...  Select detection models: These solutions focus on the automatic selection of recognition model types without using a specific default model (e.g., choosing between CNN and SVM) [13] , [14] .  ... 
doi:10.1109/access.2020.3020930 fatcat:vbmtijgeeja7xgp4cyxmbfhnke

Aerosol model selection and uncertainty modelling by adaptive MCMC technique

M. Laine, J. Tamminen
2008 Atmospheric Chemistry and Physics  
for AARJ, an Adaptive Automatic Reversible Jump MCMC for model selection and model averagingl problems with a fixed number of models M 1 , . . . , M k . 3.3.1 The algorithm 20 1.  ...  In this article the Bayesian model selection and averaging is applied to the GOMOS (ESA 2007) aerosol model selection problem.  ...  This work is done under financial support from Finnish Funding Agency for Technology and Innovation (Tekes) within the project MASI -Modelling and Simulation.  ... 
doi:10.5194/acp-8-7697-2008 fatcat:fr5jscbsqjgmlf56r26rjvrcby

Tensor Data Imputation by PARAFAC with Updated Chaotic Biases by Adam Optimizer

Pooja Choudhary, Kanwal Garg
2021 International journal of recent technology and engineering  
The idea has experimented with Netflix and traffic datasets from Guangzhou, China.  ...  The biases are created and updated by a chaotic exponential factor in Adam's optimization, which reduces the imputation error.  ...  A non-deterministic model was developed using Bayesian CP factorization, which estimated the rank of CP automatically. A robust Bayesian generative model developed in [13] for tensor factorization.  ... 
doi:10.35940/ijrte.e5291.039621 fatcat:gg52oizc3ja6xk7by4ydzniqvq

Aerosol model selection and uncertainty modelling by adaptive MCMC technique

M. Laine, J. Tamminen
2008 Atmospheric Chemistry and Physics Discussions  
The technique is based on Monte Carlo sampling and it allows model selection, calculation of model posterior probabilities and model averaging in Bayesian way.  ...  The algorithm developed here is called Adaptive Automatic Reversible Jump Markov chain Monte Carlo method (AARJ).  ...  This work is done under financial support from Finnish Funding Agency for Technology and Innovation (Tekes) within the project MASI -Modelling and Simulation. Edited by: T. Bond  ... 
doi:10.5194/acpd-8-10791-2008 fatcat:67hhg52qabbjjdp6hv762rupsy

Noise-Clustered Distant Supervision for Relation Extraction: A Nonparametric Bayesian Perspective

Qing Zhang, Houfeng Wang
2017 Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing  
To address the challenge, this work presents a novel nonparametric Bayesian formulation for the task.  ...  The essence of it is a challenging incomplete multi-label classification problem with sparse and noisy features.  ...  Motivation 2: Adaptive Latent Feature Space Selection.  ... 
doi:10.18653/v1/d17-1192 dblp:conf/emnlp/ZhangW17 fatcat:jn77hhhvlrbspnnolac5aciarq

Bayesian Matrix Completion via Adaptive Relaxed Spectral Regularization

Yang Song, Jun Zhu
2016 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Bayesian matrix completion has been studied based on a low-rank matrix factorization formulation with promising results.  ...  Our Bayesian method requires no parameter tuning and can infer the number of latent factors automatically.  ...  Conclusions and Discussions We present a novel Bayesian matrix completion method with adaptive relaxed spectral regularization.  ... 
doi:10.1609/aaai.v30i1.10231 fatcat:rgcrg6ojhnhyxl7c3df4izzf7u

Optimization of Educational Systems Using Knapsack Problem

Nasim Zandi Atashbar, Fahimeh Rahimi
2011 International Journal of Machine Learning and Computing  
Learners' knowledge level is approached through a qualitative model of the level of performance that learners exhibit with respect to the concepts which are studied and are used to adapt the lesson contents  ...  Index Terms-Adaptive learning, Bayesian networks, 0/1 Knapsack, Branch and bound algorithm, Sequencing, Elearning.  ...  Adaptive Learning The goal of this paper is to select the elements which match with the learner's characteristics and features most.  ... 
doi:10.7763/ijmlc.2012.v2.187 fatcat:pmc75k2np5dkflbfqqqymj3qou

Adaptive Mixtures of Factor Analyzers [article]

Heysem Kaya, Albert Ali Salah
2015 arXiv   pre-print
We compare the proposed algorithm with related automatic model selection algorithms on a number of benchmarks.  ...  Permitting different number of factors per mixture component, the algorithm adapts the model complexity to the data complexity.  ...  In this study, we propose a novel and adaptive model selection approach for Mixtures of Factor Analyzers.  ... 
arXiv:1507.02801v2 fatcat:4lmgh6nmrjhkvn2szr3fnmnz3e

Bayesian Sparse Factor Analysis with Kernelized Observations [article]

Carlos Sevilla-Salcedo, Alejandro Guerrero-López, Pablo M. Olmos, Vanessa Gómez-Verdejo
2021 arXiv   pre-print
Additionally, by including adequate priors, it can provide compact solutions for the kernelized observations -- based in a automatic selection of Bayesian Relevance Vectors (RVs) -- and can include feature  ...  In particular, we combine probabilistic factor analysis with what we refer to as kernelized observations, in which the model focuses on reconstructing not the data itself, but its relationship with other  ...  Automatic Bayesian Relevance Vector Selection On the basis of a full N × N kernel, with a more structured ARD prior we can achieve not only the shrinkage of the number of effective latent factors, but  ... 
arXiv:2006.00968v3 fatcat:7gte64mo6zbtxb5yy7oohpuoaq

Bayesian Singing-Voice Separation

Po-Kai Yang, Chung-Chien Hsu, Jen-Tzung Chien
2014 Zenodo  
Table 1 reports the comparison of BNMF1 and BNMF2 with adaptive basis selection and ML-NMF with fixed number of bases under SMR of 0 dB.  ...  Bayesian Factorization ML estimation is prone to find an over-trained model [1] . To improve model regularization, Bayesian approach is introduced to establish NMF for single-source separation.  ... 
doi:10.5281/zenodo.1417372 fatcat:gz4p2btsmfbjrbqbdvwv7fephm

On-line adaptation and Bayesian detection of environmental changes based on a macroscopic time evolution system

Shinji Watanabe, Atsushi Nakamura
2009 2009 IEEE International Conference on Acoustics, Speech and Signal Processing  
In addition, by incorporating a Bayesian model selection approach, we realized the simultaneous on-line adaptation and detection of environmental changes.  ...  Incremental adaptation techniques for speech recognition are aimed at adjusting acoustic models quickly and stably to such time-variant acoustic characteristics.  ...  Finally, by incorporating a Bayesian model selection approach, we realized the simultaneous on-line adaptation and detection of environmental changes.  ... 
doi:10.1109/icassp.2009.4960598 dblp:conf/icassp/WatanabeN09 fatcat:2rv4uwvyfrczbpu7gx6ypzgwxu
« Previous Showing results 1 — 15 out of 40,296 results