Filters








44,074 Hits in 5.3 sec

Emergence of Object-Selective Features in Unsupervised Feature Learning

Adam Coates, Andrej Karpathy, Andrew Y. Ng
2012 Neural Information Processing Systems  
Recent work in unsupervised feature learning has focused on the goal of discovering high-level features from unlabeled images.  ...  In this paper, we aim to test the hypothesis that unsupervised feature learning methods, provided with only unlabeled data, can learn high-level, invariant features that are sensitive to commonly-occurring  ...  As a result, it is very unclear where an "object class" begins or ends in this type of patch dataset, and less clear that a completely unsupervised learning algorithm could manage to cre-ate "object-selective  ... 
dblp:conf/nips/CoatesKN12 fatcat:2qe4q76y4vffreducefpkjdj3u

Reconstruction-based Unsupervised Feature Selection: An Embedded Approach

Jundong Li, Jiliang Tang, Huan Liu
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
In this paper, we investigate how to learn the reconstruction function from the data automatically for unsupervised feature selection, and propose a novel reconstruction-based unsupervised feature selection  ...  Recently, data reconstruction error emerged as a new criterion for unsupervised feature selection, which defines feature relevance as the capability of features to approximate original data via a reconstruction  ...  Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  ... 
doi:10.24963/ijcai.2017/300 dblp:conf/ijcai/LiTL17 fatcat:l7qkumbruncxlfjcil5nuojigq

Object-Centric Representation Learning from Unlabeled Videos [article]

Ruohan Gao, Dinesh Jayaraman, Kristen Grauman
2016 arXiv   pre-print
In this work, we explore unsupervised feature learning from unlabeled video.  ...  restriction in the scope of data relevant for learning.  ...  Object-Centric Representation Learning from Unlabeled Videos  ... 
arXiv:1612.00500v1 fatcat:p4bnb4p2ubgvjkiizpicex4sti

Selective unsupervised feature learning with Convolutional Neural Network (S-CNN)

Amir Ghaderi, Vassilis Athitsos
2016 2016 23rd International Conference on Pattern Recognition (ICPR)  
This method for unsupervised feature learning is then successfully applied to a challenging object recognition task.  ...  Selective Convolutional Neural Network (S-CNN) is a simple and fast algorithm, it introduces a new way to do unsupervised feature learning, and it provides discriminative features which generalize well  ...  In Figure 1 we show the overview of algorithm. Selective search finds the important parts of the object. Then CNN learns the features to classify those important parts.  ... 
doi:10.1109/icpr.2016.7900009 dblp:conf/icpr/GhaderiA16 fatcat:7kkko462vrepfjpxamwct6nmlm

Dimensionality Reduction: An Effective Technique for Feature Selection

Swati ASonawale, Roshani Ade
2015 International Journal of Computer Applications  
By reducing the unrelated (irrelevant) and unnecessary features related to data, or by means of effectively merging original features to produce a smaller set of feature with more discriminative control  ...  It has been observed that most of the time dataset is multidimensional and larger in size.  ...  In machine learning and data mining organizations suffers addressed problems of feature selection, especially in supervised and unsupervised models, the topic of several mechanisms [5] .  ... 
doi:10.5120/20535-2893 fatcat:f2uzmepx2fbgfi5hug5d4gzn7y

Zero-shot Feature Selection via Exploiting Semantic Knowledge [article]

Zheng Wang Department of Computer Science, University of Science, Technology Beijing
2020 arXiv   pre-print
Feature selection plays an important role in pattern recognition and machine learning systems. Supervised knowledge can significantly improve the performance.  ...  Therefore, this paper studies the problem of Zero-Shot Feature Selection, i.e., building a feature selection model that generalizes well to "unseen" concepts with limited training data of "seen" concepts  ...  FSASL [10] is an unsupervised method performing structure learning and feature selection simultaneously. [33] is the classical supervised feature selection method. 5.  ... 
arXiv:1908.03464v2 fatcat:hbcnjl4lnbexrgmbx2z6codg7m

A Survey on Unsupervised Machine Learning Algorithms for Automation, Classification and Maintenance

Memoona Khanum, Tahira Mahboob, Warda Imtiaz, Humaraia Abdul Ghafoor, Rabeea Sehar
2015 International Journal of Computer Applications  
Sparse coding learns hierarchical feature representations from raw RGB-D data in an unsupervised way by using hierarchical matching pursuit.  ...  Unsupervised learning techniques are used for learn complex, highly non-linear models with millions parameters to used large amount of unlabeled data.  ...  area of feature subset selection: unsupervised learning is quite young.  ... 
doi:10.5120/21131-4058 fatcat:aq6hlihqizcehjcajdgnx7a2ti

A Study on the Effect of Feature Selection on Malware Analysis using Machine Learning

Kehinde Oluwatoyin Babaagba, Samuel Olumide Adesanya
2019 Proceedings of the 2019 8th International Conference on Educational and Information Technology - ICEIT 2019  
In this paper, the effect of feature selection in malware detection using machine learning techniques is studied.  ...  We employ supervised and unsupervised machine learning algorithms with and without feature selection. These include both classification and clustering algorithms.  ...  Furthermore, we carried out supervised learning without feature selection. Finally, unsupervised learning without feature selection was done.  ... 
doi:10.1145/3318396.3318448 fatcat:tdep7jy24rhwvh3mbzoh7p4hom

Subitizing with Variational Autoencoders [article]

Rijnder Wever, Tom F.H. Runia
2018 arXiv   pre-print
In computer vision, it has been shown that numerosity emerges as a statistical property in neural networks during unsupervised learning from simple synthetic images.  ...  Numerosity, the number of objects in a set, is a basic property of a given visual scene.  ...  The contributions of this work are the following. We explore the emergence of visual number sense in deep networks trained in an unsupervised setting on natural images.  ... 
arXiv:1808.00257v1 fatcat:glevffunrvcdbf6yaxhtm3clq4

Symbols as Self-emergent Entities in an Optimization Process of Feature Extraction and Predictions

Peter König, Norbert Krüger
2006 Biological cybernetics  
In the mammalian cortex the early sensory processing can be characterized as feature extraction resulting in local and analogue low-level representations.  ...  Secondly, addressing a stage following early visual processing we suggest to extend the unsupervised learning scheme by including predictive processes.  ...  This allows a substantial part of feature selectivity in early visual processing to be, be understood on the basis of unsupervised learning according to a small number of objectives.  ... 
doi:10.1007/s00422-006-0050-3 pmid:16496197 fatcat:kxrsgxoqwvarfpdzv3vy44vz7a

Unsupervised Category Discovery in Images Using Sparse Neural Coding

S. Waydo, C. Koch
2007 Procedings of the British Machine Vision Conference 2007  
We present an unsupervised method for learning and recognizing object categories from unlabeled images.  ...  In recognition, this model is used in a maximum-likelihood manner to classify unseen images, and we find units emerging from learning that respond selectively to specific image categories.  ...  We thank Thomas Serre and Minjoon Kouh of MIT for providing the visual system model used here as well as assistance with its operation, and Richard Murray, Jerry Marsden, and Pietro Perona at Caltech and  ... 
doi:10.5244/c.21.102 dblp:conf/bmvc/WaydoK07 fatcat:fi3s4kz2ffbgzko66w3yhyxpme

Measures for unsupervised fuzzy-rough feature selection

Neil Mac Parthaláin, Richard Jensen, J.M. Benítez, V. Loia, F. Marcelloni
2010 International Journal of Hybrid Intelligent Systems  
In this paper, some new fuzzy-rough set-based approaches to unsupervised feature selection are proposed.  ...  For supervised learning, feature selection algorithms attempt to maximise a given function of predictive accuracy.  ...  Acknowledgement The authors would like to acknowledge the financial support for this research through The Research institute of Visual Computing -Wales (RIVIC) .  ... 
doi:10.3233/his-2010-0118 fatcat:7kl3bbnfcndsjiwt4ydjh4mbgy

Measures for Unsupervised Fuzzy-Rough Feature Selection

Neil Mac Parthaláin, Richard Jensen
2009 2009 Ninth International Conference on Intelligent Systems Design and Applications  
In this paper, some new fuzzy-rough set-based approaches to unsupervised feature selection are proposed.  ...  For supervised learning, feature selection algorithms attempt to maximise a given function of predictive accuracy.  ...  Acknowledgement The authors would like to acknowledge the financial support for this research through The Research institute of Visual Computing -Wales (RIVIC) .  ... 
doi:10.1109/isda.2009.45 dblp:conf/isda/MacParthalainJ09 fatcat:ovaugcbn35dqpdd2l26j3vw6ia

Sparse Nonlinear Feature Selection Algorithm via Local Structure Learning

Jiaye Li, Guoqiu Wen, Jiangzhang Gan, Leyuan Zhang, Shanwen Zhang
2019 Emerging Science Journal  
In this paper, we propose a new unsupervised feature selection algorithm by considering the nonlinear and similarity relationships within the data.  ...  Specifically, we use a kernel function to map each feature of the data into the kernel space.  ...  In the unsupervised feature selection algorithm of the past two years, Almusallam et al. proposed an efficient unsupervised feature selection for streaming features [8] .  ... 
doi:10.28991/esj-2019-01175 fatcat:sdicjem7rnc7tky36wqjrqnvze

Machine Learning-Based State-Of-The-Art Methods For The Classification Of RNA-Seq Data [article]

Almas Jabeen, Nadeem Ahmad, Khalid Raza
2017 bioRxiv   pre-print
In this chapter, we are going to discuss various machine learning approaches for RNA-Seq data classification and their implementation.  ...  Advancements in bioinformatics, along with developments in machine learning based classification, would provide powerful toolboxes for classifying transcriptome information available through RNA-Seq data  ...  For a Big Data feature selection process such as in case of RNA-Seq data, both supervised learning and unsupervised learning can be implemented to make decision.  ... 
doi:10.1101/120592 fatcat:frdzqa4awvbuddkp4vxmyvyo2q
« Previous Showing results 1 — 15 out of 44,074 results