Deep Belief Networks using discriminative features for phone recognition

By Abdel-rahman Mohamed, Tara N. Sainath, George E. Dahl, Bhuvana Ramabhadran, Geoffrey E. Hinton, Michael A. Picheny. 2011

314 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

C3D: Generic Features for Video Analysis (2014, background): “…availability of powerful parallel machines (GPUs, CPU clusters), together with large amounts of training data, convolutional neural networks [33] have come back providing breakthroughs on many AI problems, re-lated to text [6, 34], speech recognition [35, 36] and image based problems [15, 30].”
Low-Rank Approximations for Conditional Feedforward Computation in Deep Neural Networks (2013, background): “In recent years, deep neural networks have redefined state-o f-th -art in many application domains, notably in computer vision [11] and speech processing [14].”
Deep belief networks for predicting corporate defaults (2015, background): “Various deep learning architectures, such as deep neural networks [11], [12], [13], convolutional deep neural networks [14], and deep belief networks [15], [16], [17], [18], [19] have been applied in computer vision, automatic signal recognition, and natural language processing.”
Compressing Neural Networks with the Hashing Trick (2015, background): “…et al., 2013; Sermanet et al., 2013; Zeiler & Fergus, 2014), object detection (Girshick et al., 2014; Vinyals et al., 2014), image retrieval (Razavian et al., 2014), speech recognition (Hinton et al., 2012; Graves et al., 2013; Mohamed et al., 2011), and text representation (Mikolov et al., 2013).”
Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art (2016, background): “ieee Geoscience and remote sensinG maGazine june 2016 38 [33] A.”
Kernel methods match Deep Neural Networks on TIMIT (2014, unknown): “[14].”
Learning Spatiotemporal Features with 3D Convolutional Networks (2014, methodology): “ CPU clusters), together with large amounts of training data, convolutional neural networks [33] have come back providing breakthroughs on many AI problems, related to text [6,34], speech recognition [36,35] and image based problems [30,15]. ConvNets have been applied to the problem of human pose es”
Application of Pretrained Deep Neural Networks to Large Vocabulary Speech Recognition (2012, methodology): “We have also shown that speaker adaptive features that can be leveraged within the ANN/HMM hybrid system, by using them as the input to the neural network, as was done in [8].”
Integrative Data Analysis of Multi-Platform Cancer Data with a Multimodal Deep Learning Approach (2015, background): “First, according to the empirical studies in the deep learning literature, the number of hidden variables is usually around one tenth of that of visible variables [7], [8], [9], [10], [11], [24], [25], [26].”
Stock prediction using deep learning (2016, methodology): “Successful application in speech domain [18] has led to the idea that since speech is a time series data and stock data is also time-series so this method can be used.”
Deep Learning-Based Classification of Hyperspectral Data (2014, background): “These deep architectures have been shown to yield promising performance in many field including classification or regression tasks that involve image [26], [27], [47], language [28], and speech [29].”
Towards Bayesian Deep Learning: A Survey (2016, background): “E(v,h)) (4) RBMistrainedusing‘ContrastiveDivergence’[40]ratherthanBP.Oncetrained,RBMcaninfervorhbymarginalizing out other neurons. One can also stack layers of RBM to form a deep belief network (DBN) [76], use multiple branches of deep RBN for multimodal learning [104], or combine DBN with convoluti”
A deep convolutional neural network using heterogeneous pooling for trading acoustic invariance with phonetic confusion (2013, methodology): “We report much stronger results than [28] on the same evaluation task.”
Porting concepts from DNNs back to GMMs (2013, methodology): “Us- ing discriminative features, speaker adaptation and the combination with a discriminatively trained HMM [27] lowers the error rate to 19.3% when using a 3-gram.”
A Survey on Bayesian Deep Learning (2016, methodology): “One can also stack layers of RBM to form a deep belief network (DBN) [76], use multiple branches of deep RBN for multimodal learning [104], or combine DBN with convolutional layers to form a convolutional DBN [65].”