Recent advances in convolutional neural networks

Gu, Jiuxiang; Wang, Zhenhua; Kuen, Jason; Ma, Lihua; Shahroudy, Amir; Shuai, Bing; Liu, Ting; Wang, Xingxing; Wang, Gang; Cai, Jianfei; Chen, Tsuhan

doi:10.1016/j.patcog.2017.10.013

Cited by 4,469 publications

(2,342 citation statements)

References 203 publications

(285 reference statements)

Supporting

Mentioning

1,923

Contrasting

Unclassified

Order By: Relevance

“…However, non-saturating nonlinearities such as the rectified linear unit (ReLU) allow much faster learning than these saturating nonlinearities, particularly for models that are trained on large datasets [18]. Moreover, a number of works have shown that the performance of ReLU is better than that of sigmoid and tanh activation [39]. Thus, most of the modern studies on ConvNets use ReLU to model the output of the neurons [28], [32], [33], [34].…”

Section: Activation Functionmentioning

confidence: 99%

Deep Convolutional Neural Networks for Predominant Instrument Recognition in Polyphonic Music

Han

Kim

Lee

2017

IEEE/ACM Trans. Audio Speech Lang. Process.

155

View full text Add to dashboard Cite

Abstract-Identifying musical instruments in polyphonic music recordings is a challenging but important problem in the field of music information retrieval. It enables music search by instrument, helps recognize musical genres, or can make music transcription easier and more accurate. In this paper, we present a convolutional neural network framework for predominant instrument recognition in real-world polyphonic music. We train our network from fixed-length music excerpts with a single-labeled predominant instrument and estimate an arbitrary number of predominant instruments from an audio signal with a variable length. To obtain the audio-excerpt-wise result, we aggregate multiple outputs from sliding windows over the test audio. In doing so, we investigated two different aggregation methods: one takes the average for each instrument and the other takes the instrument-wise sum followed by normalization. In addition, we conducted extensive experiments on several important factors that affect the performance, including analysis window size, identification threshold, and activation functions for neural networks to find the optimal set of parameters. Using a dataset of 10k audio excerpts from 11 instruments for evaluation, we found that convolutional neural networks are more robust than conventional methods that exploit spectral features and source separation with support vector machines. Experimental results showed that the proposed convolutional network architecture obtained an F1 measure of 0.602 for micro and 0.503 for macro, respectively, achieving 19.6% and 16.4% in performance improvement compared with other state-of-the-art algorithms.

show abstract

Section: Activation Functionmentioning

confidence: 99%

Deep Convolutional Neural Networks for Predominant Instrument Recognition in Polyphonic Music

Han

Kim

Lee

2017

IEEE/ACM Trans. Audio Speech Lang. Process.

155

View full text Add to dashboard Cite

show abstract

“…To introduce non-linearity, the Rectified Linear Unit (ReLU) activation function was used after each convolution. It has the advantage of being resistant to the vanishing gradient problem while being simple in terms of computation and was shown to work better than sigmoid and hyperbolic tangent activation functions [34]. A square-shaped sliding window is used to scan the text-line image in the direction of the writing.…”

Section: Deep Models Based On Convolutional Recurrent Neural Networkmentioning

confidence: 99%

Transcription of Spanish Historical Handwritten Documents with Deep Neural Networks

et al. 2018

View full text Add to dashboard Cite

Abstract:The digitization of historical handwritten document images is important for the preservation of cultural heritage. Moreover, the transcription of text images obtained from digitization is necessary to provide efficient information access to the content of these documents. Handwritten Text Recognition (HTR) has become an important research topic in the areas of image and computational language processing that allows us to obtain transcriptions from text images. State-of-the-art HTR systems are, however, far from perfect. One difficulty is that they have to cope with image noise and handwriting variability. Another difficulty is the presence of a large amount of Out-Of-Vocabulary (OOV) words in ancient historical texts. A solution to this problem is to use external lexical resources, but such resources might be scarce or unavailable given the nature and the age of such documents. This work proposes a solution to avoid this limitation. It consists of associating a powerful optical recognition system that will cope with image noise and variability, with a language model based on sub-lexical units that will model OOV words. Such a language modeling approach reduces the size of the lexicon while increasing the lexicon coverage. Experiments are first conducted on the publicly available Rodrigo dataset, which contains the digitization of an ancient Spanish manuscript, with a recognizer based on Hidden Markov Models (HMMs). They show that sub-lexical units outperform word units in terms of Word Error Rate (WER), Character Error Rate (CER) and OOV word accuracy rate. This approach is then applied to deep net classifiers, namely Bi-directional Long-Short Term Memory (BLSTMs) and Convolutional Recurrent Neural Nets (CRNNs). Results show that CRNNs outperform HMMs and BLSTMs, reaching the lowest WER and CER for this image dataset and significantly improving OOV recognition.

show abstract

“…In addition to the symmetries exploited here, a typical length scale, 1=Λ QCD ∼ 10 −15 m, emerges dynamically in LQCD calculations. Consequently, there are potential advantages for a convolutional approach [82][83][84] at larger lattice volumes. Convolutional layers would again have to be customized, respecting the gauge symmetry of the problem.…”

Section: Resultsmentioning

confidence: 99%

Machine learning action parameters in lattice quantum chromodynamics

2018

View full text Add to dashboard Cite

Numerical lattice quantum chromodynamics studies of the strong interaction are important in many aspects of particle and nuclear physics. Such studies require significant computing resources to undertake. A number of proposed methods promise improved efficiency of lattice calculations, and access to regions of parameter space that are currently computationally intractable, via multi-scale action-matching approaches that necessitate parametric regression of generated lattice datasets. The applicability of machine learning to this regression task is investigated, with deep neural networks found to provide an efficient solution even in cases where approaches such as principal component analysis fail. The high information content and complex symmetries inherent in lattice QCD datasets require custom neural network layers to be introduced and present opportunities for further development.

show abstract

Recent advances in convolutional neural networks

Cited by 4,469 publications

References 203 publications

Deep Convolutional Neural Networks for Predominant Instrument Recognition in Polyphonic Music

Deep Convolutional Neural Networks for Predominant Instrument Recognition in Polyphonic Music

Transcription of Spanish Historical Handwritten Documents with Deep Neural Networks

Machine learning action parameters in lattice quantum chromodynamics

Contact Info

Product

Resources

About