Discriminative learning in sequential pattern recognition

By Xiaodong He, Li Deng, Wu Chou. 2008

202 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Speech & Audio, 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 Speech & Audio

In Signal Processing

Citations in the corpus

AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networks (2017, background): “, [11, 8]), the attention-driven image-text matching score between the entire image (Q) and the whole text description (D) is defined as”
Stacked Cross Attention for Image-Text Matching (2018, methodology): “Inspired by the minimum classification error formulation in speech recognition [18, 15], the similarity between image I and sentence T is calculated by LogSumExp pooling (LSE), i.”
Towards Better Decoding and Language Model Integration in Sequence to Sequence Models (2016, neutral): “In discriminative training of classical ASR systems normalization is performed over lattices [31].”
Deep Learning: Methods and Applications (2014, methodology): “Here, it is useful to point out a connection between the above pretraining/fine-tuning strategy associated with hybrid deep networks and the highly popular minimum phone error (MPE) training technique for the HMM (see [147, 290] for an overview).”
End-to-end Continuous Speech Recognition using Attention-based Recurrent NN: First Results (2014, methodology): “Once a per-frame classifier is trained, the whole system including the acoustic model and the HMM can be tuned jointly on full sentences to minimize the decoding error (LeCun et al., 1998; He et al., 2008; Kingsbury, 2009; Vesel et al., 2013).”
A tutorial survey of architectures, algorithms, and applications for deep learning (2014, methodology): “The framework described in (He and Deng, 2011) enables end-to-end performance optimization in the overall deep architecture using the unified learning framework initially published in (He et al., 2008).”
Optimization in speech-centric information processing: Criteria and techniques (2012, methodology): “In [4], a unified discriminative training criterion is proposed for ASR and is later extended to speech translation based on the Bayesian framework [3] and with the similar growth-transformation (GT) or EBW-based optimization method.”
Task Loss Estimation for Sequence Prediction (2015, background): “The recent survey (He et al., 2008) explains and documents improvements in speech recognition brought by other methods of discriminative training of speech recognition systems.”
Convolutional Neural Networks for Speech Recognition (2014, background): “Gaussian mixture models (GMMs) have been, until very recently, regarded as the most powerful model Manuscript received October 11, 2013; revised February 04, 2014; accepted July 05, 2014.”
Machine Learning Paradigms for Speech Recognition: An Overview (2013, unknown): “The maximum mutual information estimation (MMIE) of generative models, highly popular in ASR, uses an equivalent loss function to the conditional likelihood loss that leads to the empirical risk of (33) See a simple proof of their equivalence in [74].”
Deep Neural Networks for Acoustic Modeling in Speech Recognition (2012, background): “…sequence classification criteria, which are more directly correlated with the overall word or phone error rate, can be very helpful in improving recognition accuracy [7], [35] and the benefit of GRBM RBM RBM DBN DBN-DNN Copy Copy W 1 W 2 W 3 W 3 W 4 = 0 W 2 W 1 W 3 T W 2 T W 1 T [fig1] the sequence”
Use of kernel deep convex networks and end-to-end learning for spoken language understanding (2012, methodology): “Therefore, the efficient EBW-based optimization method (He et al. 2008, He and Deng 2012) is not suitable anymore.”
Visual Attention Methods in Deep Learning: An In-Depth Survey (2022, methodology): “…a) compute cosine similarity for all image-text pairs [82] followed by ℓ 2 normalization [169], b) compute the weighted sum of these pairs attentions, where the image one is calculated by softmax [26], c) the final similarity between these pairs is computed using LogSumExp pooling [52, 73].”