Towards Two-Dimensional Sequence to Sequence Model in Neural Machine Translation

Bahar, Parnia; Brix, Christopher; Ney, Hermann

doi:10.48550/arxiv.1810.03975

Cited by 3 publications

(12 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As given in Fig. 5, 2D LSTM network can be realized when the LSTM cell is recursively in a 2D mesh form [13]. Each LSTM cell utilizes the hidden and cell states from the two neighboring cells in the left and below positions in the mesh.…”

Section: B Framework Architecture and Methodsmentioning

confidence: 99%

Applying Deep-Learning-Based Computer Vision to Wireless Communications: Methodologies, Opportunities, and Challenges

Tian¹,

Pan²,

Alouini³

2020

Preprint

View full text Add to dashboard Cite

Section: B Framework Architecture and Methodsmentioning

confidence: 99%

Applying Deep-Learning-Based Computer Vision to Wireless Communications: Methodologies, Opportunities, and Challenges

Tian¹,

Pan²,

Alouini³

2020

Preprint

View full text Add to dashboard Cite

“…1: The internal architecture of the standard and the 2DLSTM. The additional connections are marked in blue [13]. matic extraction of features from raw 2D-images over convolutional neural networks (CNNs) [14].…”

Section: Related Workmentioning

confidence: 99%

“…Recently, the 2DLSTM layer also has been used for sequence-to-sequence modeling in machine translation [13] where it implicitly updates the source representation conditioned on the generated target words. In a similar direction, a 2D CNN-based network has been proposed where the positions of the source and the target words define the 2D grid for translation modeling [21].…”

Section: Related Workmentioning

confidence: 99%

“…Similar to [13], we apply a 2DLSTM layer to combine the acoustic model (the LSTM encoder) and the language model (the decoder) without any attention components. The 2DL-STM reconciles the context from both the input and the output sequences and re-interprets the encoder states while a new word has been predicted.…”

Section: Related Workmentioning

confidence: 99%

“…In contrast to the attention-based sequence-tosequence model, where the encoder states are not updated and the model is not able to re-interpret the encoder states while decoding, this model enables the computation of the encoding of the observation sequence as a function of the previously generated transcribed words. Our model is similar to an architecture used in machine translation described in [13]. We believe that the 2DLSTM is able to capture necessary monotonic alignments as well as retrieve coverage concepts internally by its cell states.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On Using 2D Sequence-to-sequence Models for Speech Recognition

Bahar

Zeyer

Schlüter

et al. 2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Attention-based sequence-to-sequence models have shown promising results in automatic speech recognition. Using these architectures, one-dimensional input and output sequences are related by an attention approach, thereby replacing more explicit alignment processes, like in classical HMM-based modeling. In contrast, here we apply a novel two-dimensional long short-term memory (2DLSTM) architecture to directly model the input/output relation between audio/feature vector sequences and word sequences. The proposed model is an alternative model such that instead of using any type of attention components, we apply a 2DLSTM layer to assimilate the context from both input observations and output transcriptions. The experimental evaluation on the Switchboard 300h automatic speech recognition task shows word error rates for the 2DLSTM model that are competitive to end-to-end attention-based model.

show abstract

Viewing Channel as Sequence Rather Than Image: A 2-D Seq2Seq Approach for Efficient MIMO-OFDM CSI Feedback

Chen

Zhang

Xiao

et al. 2023

IEEE Trans. Wireless Commun.

View full text Add to dashboard Cite

In this paper, we aim to design an effective learning-based channel state information (CSI) feedback scheme for the multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems from a physics-inspired perspective. We first argue that the CSI matrix of a MIMO-OFDM system is in general physically closer to a two-dimensional (2-D) sequence rather than an image, due to its apparent unsmoothness, non-scalablity and translational variance within both the spatial and frequency domains. On this basis, we introduce a 2-D long short-term memory (LSTM) neural network to represent the CSI and propose a 2-D sequence-to-sequence (Seq2Seq) model for CSI compression and reconstruction. Specifically, one two-layer 2-D LSTM is used for CSI feature extraction, and another two-layer 2-D LSTM is used for CSI representation and reconstruction. The proposed scheme can not only fully utilize the unique 2-D characteristics of CSI, but also well preserve the index information and unsmooth features of the CSI matrix compared with current convolutional neural network (CNN) based schemes. We show the computational complexity of the proposed scheme is linear in the number of transmit antennas and subcarriers. Its key performances like reconstruction accuracy, convergence speed, generalization ability after short-term training, as well as robustness to lossy feedback are comprehensively compared with existing popular convolutional networks. Experiment results show that our scheme can bring up to nearly 7 dB gain in reconstruction accuracy under the same overhead and reduce feedback overhead by up to 75% under the same accuracy compared with the conventional CNN-based approaches.

show abstract

Towards Two-Dimensional Sequence to Sequence Model in Neural Machine Translation

Cited by 3 publications

References 0 publications

Applying Deep-Learning-Based Computer Vision to Wireless Communications: Methodologies, Opportunities, and Challenges

Applying Deep-Learning-Based Computer Vision to Wireless Communications: Methodologies, Opportunities, and Challenges

On Using 2D Sequence-to-sequence Models for Speech Recognition

Viewing Channel as Sequence Rather Than Image: A 2-D Seq2Seq Approach for Efficient MIMO-OFDM CSI Feedback

Contact Info

Product

Resources

About