Results 1 - 10
of
74
Feature engineering in context-dependent deep neural networks for conversational speech transcription
- in ASRU
, 2011
"... Abstract—We investigate the potential of Context-Dependent Deep-Neural-Network HMMs, or CD-DNN-HMMs, from a feature-engineering perspective. Recently, we had shown that for speaker-independent transcription of phone calls (NIST RT03S Fisher data), CD-DNN-HMMs reduced the word error rate by as much a ..."
Abstract
-
Cited by 76 (15 self)
- Add to MetaCart
Abstract—We investigate the potential of Context-Dependent Deep-Neural-Network HMMs, or CD-DNN-HMMs, from a feature-engineering perspective. Recently, we had shown that for speaker-independent transcription of phone calls (NIST RT03S Fisher data), CD-DNN-HMMs reduced the word error rate by as much
Simultaneous feature learning and hash coding with deep neural networks
- in Proc. IEEE Conference on Computer Vision and Pattern Recognition
, 2015
"... Similarity-preserving hashing is a widely-used method for nearest neighbour search in large-scale image retrieval tasks. For most existing hashing methods, an image is first encoded as a vector of hand-engineering visual fea-tures, followed by another separate projection or quantiza-tion step that g ..."
Abstract
-
Cited by 4 (0 self)
- Add to MetaCart
neural networks. The pipeline of the proposed deep architecture consists of three building blocks: 1) a sub-network with a stack of convolution lay-ers to produce the effective intermediate image features; 2) a divide-and-encode module to divide the intermediate im-age features into multiple branches
Quadratic Features and Deep Architectures for Chunking
"... We experiment with several chunking models. Deeper architectures achieve better generalization. Quadratic filters, a simplification of a theoretical model of V1 complex cells, reliably increase accuracy. In fact, logistic regression with quadratic filters outperforms a standard single hidden layer n ..."
Abstract
-
Cited by 1 (1 self)
- Add to MetaCart
neural network. Adding quadratic filters to logistic regression is almost as effective as feature engineering. Despite predicting each output label independently, our model is competitive with ones that use previous decisions. 1
Deep Neural Networks for Named Entity Recognition in Italian
"... English. In this paper, we intro-duce a Deep Neural Network (DNN) for engineering Named Entity Recognizers (NERs) in Italian. Our network uses a sliding window of word contexts to pre-dict tags. It relies on a simple word-level log-likelihood as a cost function and uses a new recurrent feedback mech ..."
Abstract
- Add to MetaCart
English. In this paper, we intro-duce a Deep Neural Network (DNN) for engineering Named Entity Recognizers (NERs) in Italian. Our network uses a sliding window of word contexts to pre-dict tags. It relies on a simple word-level log-likelihood as a cost function and uses a new recurrent feedback
Applications of Artificial Neural Networks in Foundation Engineering
"... Abstract: In recent years, artificial neural networks (ANNs) have emerged as one of the potentially most successful modelling approaches in engineering. In particular, ANNs have been applied to many areas of geotechnical engineering and have demonstrated considerable success. The objective of this p ..."
Abstract
- Add to MetaCart
of this paper is to highlight the use of ANNs in foundation engineering. The paper describes ANN techniques and some of their applications in shallow and deep foundations, as well as the salient features associated with ANN model development. Finally, the paper discusses the strengths and limitations of ANNs
DEEP-CARVING: Discovering Visual Attributes by Carving Deep Neural Nets
"... Most of the approaches for discovering visual attributes in images demand significant supervision, which is cumbersome to obtain. In this paper, we aim to discover visual attributes in a weakly supervised setting that is commonly encountered with contemporary image search engines. For instance, give ..."
Abstract
- Add to MetaCart
the attributes present in a test image. Deep Convolutional Neural Networks (CNNs) [20] have en-joyed remarkable success in vision applications recently. How-ever, in a weakly supervised scenario, widely used CNN train-ing procedures do not learn a robust model for predicting mul-tiple attribute labels
Machining Quality Predictions: Comparative Analysis of Neural Network and Fuzzy Logic
"... Abstract — Surface finish is an important objective function in manufacturing engineering. It holds the characteristic that could influence the performance of mechanical parts which is also proportional to production cost. It is also an aspect for designing mechanical elements and frequently present ..."
Abstract
- Add to MetaCart
work. From the findings, it is found that Sugeno Fuzzy model gives better the closest values as compared to the ANN model. Thus, the work conditions and Fuzzy environment is selected for predictions of surface roughness in drilling. Index Term—Sugeno Fuzzy, deep drilling, neural network, fuzzy logic
Machining Quality Predictions: Comparative Analysis of Neural Network and Fuzzy Logic
"... Abstract — Surface finish is an important objective function in manufacturing engineering. It holds the characteristic that could influence the performance of mechanical parts which is also proportional to production cost. It is also an aspect for designing mechanical elements and frequently present ..."
Abstract
- Add to MetaCart
work. From the findings, it is found that Sugeno Fuzzy model gives better the closest values as compared to the ANN model. Thus, the work conditions and Fuzzy environment is selected for predictions of surface roughness in drilling. Index Terms—Sugeno Fuzzy, deep drilling, neural network, fuzzy logic
Enhanced Higgs to τ+τ − Search with Deep Learning
"... The Higgs boson is thought to provide the interaction that imparts mass to the fundamental fermions, but while measurements at the Large Hadron Collider (LHC) are consistent with this hy-pothesis, current analysis techniques lack the statistical power to cross the traditional 5σ significance barrier ..."
Abstract
-
Cited by 1 (1 self)
- Add to MetaCart
barrier without more data. Deep learning techniques have the potential to increase the statistical power of this analysis by automatically learning complex, high-level data representations. In this work, deep neural networks are used to detect the decay of the Higgs to a pair of tau leptons. A Bayesian
On the Recursive Neural Networks for Relation Extraction and Entity Recognition
"... Recently there has been a surge of interest in neural architectures for complex structured learning tasks. Along this track, we are ad-dressing the supervised task of relation extrac-tion and named-entity recognition via recur-sive neural structures and deep unsupervised feature learning. Our models ..."
Abstract
- Add to MetaCart
Recently there has been a surge of interest in neural architectures for complex structured learning tasks. Along this track, we are ad-dressing the supervised task of relation extrac-tion and named-entity recognition via recur-sive neural structures and deep unsupervised feature learning. Our
Results 1 - 10
of
74