paper-with-me

홈 › Papers

Dynamic Multi-Branch Layers for On-Device Neural Machine Translation

2021-05-14 · Zhixing Tan, Zeyuan Yang, Meng Zhang, Qun Liu, Maosong Sun, Yang Liu

With the rapid development of artificial intelligence (AI), there is a trend in moving AI applications, such as neural machine translation (NMT), from cloud to mobile devices. Constrained by limited hardware resources and battery, the performance of on-device NMT systems is far from satisfactory. Inspired by conditional computation, we propose to improve the performance of on-device NMT systems with dynamic multi-branch layers. Specifically, we design a layer-wise dynamic multi-branch network with only one branch activated during training and inference. As not all branches are activated during training, we propose shared-private reparameterization to ensure sufficient training for each branch. At almost the same computational cost, our method achieves improvements of up to 1.7 BLEU points on the WMT14 English-German translation task and 1.8 BLEU points on the WMT20 Chinese-English translation task over the Transformer model, respectively. Compared with a strong baseline that also uses multiple branches, the proposed method is up to 1.5 times faster with the same number of parameters.

📄 PDF Abstract BibTeX arXiv:2105.06679

Code (1)

thunlp-mt/transformer-dmb 공식 구현 pytorch

Tasks

Machine TranslationNMTTranslation

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Residual Connection 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Adam 설명 없음

Similar Papers 제목 키워드 기반

Splitformer: An improved early-exit architecture for automatic speech recognition on edge devices

2025-06-22 · Maxence Lasbordes, Daniele Falavigna, Alessio Brutti

The ability to dynamically adjust the computational load of neural models during inference in a resource aware manner is crucial for on-device processing scenarios, characterised by limited and time-varying computational…

Automatic Speech Recognitionspeech-recognitionSpeech Recognition

Dynamic Perceiver for Efficient Visual Recognition

2023-06-20 · ICCV 2023 1 · Yizeng Han, Dongchen Han, Zeyu Liu, Yulin Wang 외

Early exiting has become a promising approach to improving the inference efficiency of deep networks. By structuring models with multiple classifiers (exits), predictions for ``easy'' samples can be generated at earlier …

Action RecognitionClassificationCPUEarly Classification+5

Dynamic Switch Layers For Unsupervised Learning

2024-04-05 · Haiguang Li, Usama Pervaiz, Michał Matuszak, Robert Kamara 외

On-device machine learning (ODML) enables intelligent applications on resource-constrained devices. However, power consumption poses a major challenge, forcing a trade-off between model accuracy and power efficiency that…

Multi-branch Spatio-Temporal Graph Neural Network For Efficient Ice Layer Thickness Prediction

2024-11-06 · Zesheng Liu, Maryam Rahnemoonfar

Understanding spatio-temporal patterns in polar ice layers is essential for tracking changes in ice sheet balance and assessing ice dynamics. While convolutional neural networks are widely used in learning ice layer patt…

Graph Neural Network

MABNet: A Lightweight Stereo Network Based on Multibranch Adjustable Bottleneck Module

2020-08-01 · ECCV 2020 8 · Jiabin Xing, Zhi Qi, Jiying Dong, Jiaxuan Cai 외

Recently, end-to-end CNNs have presented remarkable performance for disparity estimation. But most of them are too heavy to resource-constrained devices, because of enormous parameters necessary for satisfactory results.…

Disparity Estimation