Unconstrained On-line Handwriting Recognition with Recurrent Neural Networks

By Alex Graves, Santiago Fernández, Marcus Liwicki, Horst Bunke, J. Urgen Schmidhuber. 2007

254 citations recorded by the source corpus.

How researchers see this paper

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

In Natural Language Processing

Citations in the corpus

Speech recognition with deep recurrent neural networks (2013, background): “The combination of Long Short-term Memory [11], an RNN architecture with an improved memory, with end-to-end training has proved especially effective for cursive handwriting recognition [12, 13].”
The Goldilocks Principle: Reading Children's Books with Explicit Memory Representations (2015, methodology): “The attentive/impatient reading models encode the articles using bidirectional RNNs (Graves et al., 2008).”
Sequence Transduction with Recurrent Neural Networks (2012, background): “, 2011), or using delayed strokes to identify handwritten characters from pen trajectories (Graves et al., 2008).”
Towards Universal Paraphrastic Sentence Embeddings (2015, background): “Some have been based on distributional semantics (Baroni et al., 2014; Paperno et al., 2014; Polajnar et al., 2015; Tian et al., 2015), while the current trend is toward development of neural network architectures.”
LSTM: A Search Space Odyssey (2015, methodology): “[45] using a different and much more extensive preprocessing.”
Hierarchical Multiscale Recurrent Neural Networks (2016, background): “r spatial data (Krizhevsky et al., 2012; Mnih et al., 2015; Silver et al., 2016). For modelling temporal data, the recent resurgence of recurrent neural networks (RNN) has led to remarkable advances (Graves et al., 2008, 2013; Cho et al., 2014; Sutskever et al., 2014; Vinyals et al., 2015). However,”
Deep learning in neural networks: An overview (2014, methodology): “mportant applications of LSTM include protein analysis (Hochreiter and Obermayer, 2005), robot localization (Forster et al., 2007) and¨ 17 robot control (Mayer et al., 2008), handwriting recognition (Graves et al., 2008, 2009; Graves and Schmidhuber, 2009; Bluche et al., 2014), optical character rec”
Stochastic Language Generation in Dialogue using Recurrent Neural Networks with Convolutional Sentence Reranking (2015, background): “Previously, bidirectional RNNs (Schuster and Paliwal, 1997) have been shown to be effective for handwriting recognition (Graves et al., 2008), speech recognition”
Learning Stochastic Recurrent Networks (2014, background): “1 Introduction Recurrent Neural Networks (RNNs) are ﬂexible and powerful tools for modeling sequences. While only bearing marginal existence in the 1990’s, recent successes in real world applications [1, 2, 3, 4, 5] have resurged interest. This is partially due to architectural enhancements [6], new”
Self-Delimiting Neural Networks (2012, background): “d problems with learning to store relevant events in short-term memory across long time lags [26]. Long Short-Term Memory (LSTM) overcame these problems, outperforming early RNNs in many applications [27, 14, 15, 65, 20, 21, 23, 22]. While RNNs used to be toy problem methods in the 1990s, they have ”
From speech to letters - using a novel neural network architecture for grapheme based ASR (2009, methodology): “A more detailed description of the CTC algorithm can be found in [3].”
Deep learning helicopter dynamics models (2015, background): “Recently, Deep Learning (regression using hierarchical neural network representations) has had much success in high-dimensional learning problems in domains like image understanding [18], natural language processing [19, 20], and most relevantly with time-series inputs as in handwriting [21] and spe”
Improving Offline Handwritten Text Recognition with Hybrid HMM/ANN Models (2011, methodology): “our own and others’ experience in using hybrid HMM/ANN models for automatic speech recognition [13], [14], [15], [16], [17] and for online handwriting recognition [18], [19], [20], [21], [22], [23], .”
ReNet: A Recurrent Neural Network Based Alternative to Convolutional Networks (2015, background): “Other popular areas of application include ofﬂine handwriting recognition/generation [Graves and Schmidhuber, 2009, Graves et al., 2008, Graves, 2013] and speech recognition [Chorowski et al., 2014, Graves and Jaitly, 2014].”
Recent Advances in Recurrent Neural Networks (2017, methodology): “RNNs perform well on this task, given the contextual dependencies in letter sequences [122].”
Data-driven forecasting of high-dimensional chaotic systems with long short-term memory networks (2018, methodology): “gely to the LSTM, that copes eectively with the aforementioned problem using gates. The LSTM has been successfully applied in sequence modeling [41], speech recognition [39], handwriting recognition [40] and language translation [42]. The equations of the LSTM are gf t= ˙ f W [h 1; ] + b gi t= ˙ i W”
Personal recommendation using deep recurrent neural networks in NetEase (2016, methodology): “It becomes an effective tool for processing nonlinear time series applications [10][17], such as handwriting recognition [6] and speech recognition [7].”
Reinforcement Learning for Primary care e Appointment Scheduling (2017, background): “They extend the notion of typical FFN by adding inter-layer and self-connections to units in the recurrent layer [43].”