A deep architecture with bilinear modeling of hidden representations: Applications to phonetic recognition

By Brian Hutchinson, Li Deng, Dong Yu. 2012

48 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Mathematics, 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 Computational Mathematics

In Signal Processing

Citations in the corpus

Deep Learning: Methods and Applications (2014, methodology): “The above DSN architecture has recently been generalized to its tensorized version, which we call the tensor DSN (TDSN) [180, 181].”
A tutorial survey of architectures, algorithms, and applications for deep learning (2014, background): “That is, modules of the TDSN are stacking up in a similar way to form a deep architecture.”
Scalable stacking and learning for building deep architectures (2012, background): “Further, our recent extension of the DSN architecture to a tensor version [15] has demonstrated huge architectural flexibilty of the stacking and parallel learning mechanisms presented in this paper, and a vast opportunity exists in exploiting such flexibility for potential successful applications i”
Acoustic Modeling With Hierarchical Reservoirs (2013, background): “Due to this, alternative deep architectures like Deep Convex Networks [44] or Deep Stacking Networks [12], [13] were proposed recently to overcome the scalability challenge of DBNs by solving convex learning problems in each module, allowing a faster parallel batch training.”
Parallel Training for Deep Stacking Networks (2012, background): “[3], and will not be covered in this paper.”
Deep Neural Networks for Acoustic Modeling in Speech Recognition (2012, background): “Some recent attempts are described in [52] and [57].”
On building ensembles of stacked denoising auto-encoding classifiers and their further improvement (2018, background): “On the other hand, another sequentially trainable deep architecture, the Deep Stacking Networks (DSNs), was introduced in [13,14] , following the idea of training shallow MLPs and adding their outputs to the input vector for training further units.”
Large Vocabulary Speech Recognition Using Deep Tensor Neural Networks (2012, background): “This suggests that the DP (and tensor) layers are better suited to operate upon binary features, consistent with the findings from [13].”
Ensemble deep learning: A review (2021, background): “Comprehensive evaluation and the more detailed analysis of the learning algorithm and T-DSN implementation was given in [85].”
Machine Learning Paradigms for Speech Recognition: An Overview (2013, background): “Examples include deep-structured CRF, tandem-MLP architecture [94], [258], deep convex or stacking network [248] and its tensor version [242], [243], [259], and detection-based ASR architecture [235], [236], [260].”
Use of kernel deep convex networks and end-to-end learning for spoken language understanding (2012, background): “In a recent extension of DCN, a tensor structure was imposed, shifting the majority of non-convex learning burden for into a convex one (Hutchinson et al, 2012).”
The Deep Tensor Neural Network With Applications to Large Vocabulary Speech Recognition (2013, methodology): “More recent work [20] replaced the single sigmoid hidden layer with a tensor layer in a deep network where blocks of shallow networks are used to construct the stacking deep architecture and each block in the stacking network consists of only one hidden layer.”