Filters








62,335 Hits in 5.2 sec

Building high-level features using large scale unsupervised learning [article]

Quoc V. Le, Marc'Aurelio Ranzato, Rajat Monga, Matthieu Devin, Kai Chen, Greg S. Corrado, Jeff Dean, Andrew Y. Ng
2012 arXiv   pre-print
We consider the problem of building high-level, class-specific feature detectors from only unlabeled data. For example, is it possible to learn a face detector using only unlabeled images?  ...  We also find that the same network is sensitive to other high-level concepts such as cat faces and human bodies.  ...  Introduction The focus of this work is to build high-level, classspecific feature detectors from unlabeled images.  ... 
arXiv:1112.6209v5 fatcat:67htqal3pzeexoeuvd7guxtea4

Building high-level features using large scale unsupervised learning

Quoc V. Le
2013 2013 IEEE International Conference on Acoustics, Speech and Signal Processing  
We consider the problem of building highlevel, class-specific feature detectors from only unlabeled data. For example, is it possible to learn a face detector using only unlabeled images?  ...  We also find that the same network is sensitive to other high-level concepts such as cat faces and human bodies.  ...  Introduction The focus of this work is to build high-level, classspecific feature detectors from unlabeled images.  ... 
doi:10.1109/icassp.2013.6639343 dblp:conf/icassp/Le13 fatcat:kusvpcqjcra7napchs7e3kt4hu

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  
Deep belief networks (DBNs) and sparse coding are the two well known techniques of unsupervised learning models. For large scale application  ...  Unsupervised learning techniques are used for learn complex, highly non-linear models with millions parameters to used large amount of unlabeled data.  ...  Unsupervised feature learning and deep learning have surfaced as methodologies in machine learning from unlabeled data. 2.12)Building High-level Features Using Large Scale CONCLUSION Research in the  ... 
doi:10.5120/21131-4058 fatcat:aq6hlihqizcehjcajdgnx7a2ti

Hierarchical Deep Learning Architecture For 10K Objects Classification [article]

Atul Laxman Katole, Krishna Prasad Yellapragada, Amish Kumar Bedi, Sehaj Singh Kalra, Mynepalli Siva Chaitanya
2015 arXiv   pre-print
Both the approaches yet cannot scale up realistically to provide recognition for a very large number of objects as high as 10K.  ...  These architectures extract & learn the real world hierarchical visual features utilizing supervised & unsupervised learning approaches respectively.  ...  They learn high level abstractions from low level features extracted from images utilizing supervised or unsupervised learning algorithms.  ... 
arXiv:1509.01951v1 fatcat:zm6636eulfgcpgn3qgai36tpke

Deep learning applications and challenges in big data analytics

Maryam M Najafabadi, Flavio Villanustre, Taghi M Khoshgoftaar, Naeem Seliya, Randall Wald, Edin Muharemagic
2015 Journal of Big Data  
Deep Learning algorithms extract high-level, complex abstractions as data representations through a hierarchical learning process.  ...  A key benefit of Deep Learning is the analysis and learning of massive amounts of unsupervised data, making it a valuable tool for Big Data Analytics where raw data is largely unlabeled and un-categorized  ...  Their work was a large scale investigation on the feasibility of building high-level features with Deep Learning using only unlabeled (unsupervised) data, and clearly demonstrated the benefits of using  ... 
doi:10.1186/s40537-014-0007-7 fatcat:65mi6dnv5rg6poesotupqbsm7y

Unsupervised Learning for Satellite Image Classification

Giriraja C.V, Srinivasa C, T.K. Jaya Ram, Avula Haswanth
2014 IOSR Journal of VLSI and Signal processing  
., with the help of Support Vector Machine (SVM) and unsupervised learning method using MATLAB is presented.  ...  Dense low-level feature descriptors are extracted and exploited in a novel way to learn a set of basis functions.  ...  Detection system based on the proposed feature extraction and learning approaches for detecting largefacility in large-scale high-resolution aerial imagery.  ... 
doi:10.9790/4200-04240104 fatcat:pj5vigcrfnecbfrp2ryne3ugka

Disentangled Latent Transformer for Interpretable Monocular Height Estimation [article]

Zhitong Xiong, Sining Chen, Yilei Shi, Xiao Xiang Zhu
2022 arXiv   pre-print
Exploring the semantic and height selectivity of the learned internal deep representations; 2) Instances: object-level interpretation.  ...  Studying the effects of different semantic classes, scales, and spatial contexts on height estimation; 3) Attribution: pixel-level analysis.  ...  Building Ground Water Others High Vege. Building Ground High Vege. Building Water Others Ground High Vege.  ... 
arXiv:2201.06357v2 fatcat:g3zpj7pr2fg2jhkau6t52wvaqy

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.  ...  Here, we propose a large-scale feature learning system that enables us to carry out this experiment, learning 150,000 features from tens of millions of unlabeled images.  ...  For this purpose we use the K-means-like method used by [2] , which has previously been used for large-scale feature learning.  ... 
dblp:conf/nips/CoatesKN12 fatcat:2qe4q76y4vffreducefpkjdj3u

Machine Learning with Big Data

Amit Kumar Tyagi, Rekha G
2019 Social Science Research Network  
Deep learning algorithms extract high-level, complex abstractions as data representations through a hierarchical learning process.  ...  While deep learning can be applied to learn from labeled data, it is primarily attractive for learning from large amounts of unlabeled/unsupervised data, making it attractive for extracting meaningful  ...  Deep learning challenges in big data analytics lie in: incremental learning for non-stationary data, high-dimensional data, and large-scale models [13] .  ... 
doi:10.2139/ssrn.3356269 fatcat:m7ehu6uh45hitczq5gn4i5rtay

Unsupervised Single-Image Super-Resolution with Multi-Gram Loss

Yong Shi, Biao Li, Bo Wang, Zhiquan Qi, Jiabin Liu
2019 Electronics  
In this paper, we propose a novel unsupervised method named unsupervised single-image SR with multi-gram loss (UMGSR) to overcome the dilemma.  ...  However, most methods train in the dataset with a fixed kernel (such as bicubic) between high-resolution images and their low-resolution counterparts.  ...  Pixel-level loss. Pixel-level loss is used to recover high-frequency information in I SR i with supervised I HR i .  ... 
doi:10.3390/electronics8080833 fatcat:thbnxnljgbc25ktechw5w2yiam

Unsupervised Learning of Semantics of Object Detections for Scene Categorization [chapter]

Grégoire Mesnil, Salah Rifai, Antoine Bordes, Xavier Glorot, Yoshua Bengio, Pascal Vincent
2014 Advances in Intelligent Systems and Computing  
These models define high-level representations by combining semantic lower-level elements, e.g., detection of object parts.  ...  We also show that the uncertainty relative to object detectors hampers the use of external semantic knowledge to improve detectors combination, unlike our unsupervised learning approach. 210 G.  ...  Codes for the experiments have been implemented using Theano [4] Machine Learning library.  ... 
doi:10.1007/978-3-319-12610-4_13 fatcat:a36rlagtizg4levkrxozjaiehq

"Burn-in, bias, and the rationality of anchoring"

Falk Lieder, Thomas L. Griffiths, Noah D. Goodman
2012 Neural Information Processing Systems  
Recent work in unsupervised feature learning has focused on the goal of discovering high-level features from unlabeled images.  ...  Here, we propose a large-scale feature learning system that enables us to carry out this experiment, learning 150,000 features from tens of millions of unlabeled images.  ...  For this purpose we use the K-means-like method used by [2] , which has previously been used for large-scale feature learning.  ... 
dblp:conf/nips/LiederGG12 fatcat:74t72gjdizhvxe45sspjmb54vq

Adversarial Learning based Discriminative Domain Adaptation for Geospatial Image Analysis

Nikhil Makkar, Hsiuhan Lexie Yang, Saurabh Prasad
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Multisensor high-resolution images from National Agriculture Imagery Program and SpaceNet-Rio datasets were used as the source and target for the task of building extraction for large scale semantic segmentation  ...  We are using adversarial learning to extract discriminative target domain features that are aligned with source domain.  ...  Second, another task-specific ADDA framework is presented for large scale domain adaptation for semantic segmentation applications for high-resolution satellite imagery, and we use building extraction  ... 
doi:10.1109/jstars.2021.3132259 fatcat:5ppi25cwirc2bmnlgolauiwga4

Machine Learning in Big Data

Lidong Wang, Cheryl Ann Alexander
2016 International journal of mathematical, engineering and management sciences  
This paper introduces methods in machine learning, main technologies in Big Data, and some applications of machine learning in Big Data.  ...  Challenges of machine learning applications in Big Data are discussed. Some new methods and technology progress of machine learning in Big Data are also presented.  ...  Deep learning challenges in big data analytics lie in: incremental learning for non-stationary data, high-dimensional data, and large-scale models (Najafabadi et al., 2015) .  ... 
doi:10.33889/ijmems.2016.1.2-006 fatcat:eidif7z3afbihflemxwcnxo7xi

Leveraging Stacked Denoising Autoencoder in Prediction of Pathogen-Host Protein-Protein Interactions

Huaming Chen, Jun Shen, Lei Wang, Jiangning Song
2017 2017 IEEE International Congress on Big Data (BigData Congress)  
In this paper, we further detail the framework based on unsupervised learning model for PHPPI researches, while curating a large imbalanced PHPPI dataset.  ...  In this paper, we further detail the framework based on unsupervised learning model for PHPPI researches, while curating a large imbalanced PHPPI dataset.  ...  These successes in data feature representation motivate us to introduce unsupervised learning model into our study on PHPPI.  ... 
doi:10.1109/bigdatacongress.2017.54 dblp:conf/bigdata/ChenSWS17 fatcat:7dtx36evw5hyjhe5tp2yzmo4vm
« Previous Showing results 1 — 15 out of 62,335 results