paper-with-me

홈 › Papers

Improving Deep Transformer with Depth-Scaled Initialization and Merged Attention

2019-08-29 · IJCNLP 2019 11 · Biao Zhang, Ivan Titov, Rico Sennrich

The general trend in NLP is towards increasing model capacity and performance via deeper neural networks. However, simply stacking more layers of the popular Transformer architecture for machine translation results in poor convergence and high computational overhead. Our empirical analysis suggests that convergence is poor due to gradient vanishing caused by the interaction between residual connections and layer normalization. We propose depth-scaled initialization (DS-Init), which decreases parameter variance at the initialization stage, and reduces output variance of residual connections so as to ease gradient back-propagation through normalization layers. To address computational cost, we propose a merged attention sublayer (MAtt) which combines a simplified averagebased self-attention sublayer and the encoderdecoder attention sublayer on the decoder side. Results on WMT and IWSLT translation tasks with five translation directions show that deep Transformers with DS-Init and MAtt can substantially outperform their base counterpart in terms of BLEU (+1.1 BLEU on average for 12-layer models), while matching the decoding speed of the baseline model thanks to the efficiency improvements of MAtt.

📄 PDF Abstract BibTeX arXiv:1908.11365

Code (1)

bzhangGo/zero 공식 구현 tf

Tasks

DecoderMachine TranslationTranslation

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…
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
Residual Connection 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
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…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

Similar Papers 제목 키워드 기반

LipsFormer: Introducing Lipschitz Continuity to Vision Transformers

2023-04-19 · Xianbiao Qi, Jianan Wang, Yihao Chen, Yukai Shi 외

We present a Lipschitz continuous Transformer, called LipsFormer, to pursue training stability both theoretically and empirically for Transformer-based models. In contrast to previous practical tricks that address traini…

Subcritical Signal Propagation at Initialization in Normalization-Free Transformers

2026-04-13 · Sergey Alekseev arxiv

We study signal propagation at initialization in transformers through the averaged partial Jacobian norm (APJN), a measure of gradient amplification across layers. We extend APJN analysis to transformers with bidirection…

Infinite Limits of Multi-head Transformer Dynamics

2024-05-24 · Blake Bordelon, Hamza Tahir Chaudhry, Cengiz Pehlevan

In this work, we analyze various scaling limits of the training dynamics of transformer models in the feature learning regime. We identify the set of parameterizations that admit well-defined infinite width and depth lim…

Weight subcloning: direct initialization of transformers using larger pretrained ones

2023-12-14 · Mohammad Samragh, Mehrdad Farajtabar, Sachin Mehta, Raviteja Vemulapalli 외

Training large transformer models from scratch for a target task requires lots of data and is computationally demanding. The usual practice of transfer learning overcomes this challenge by initializing the model with wei…

image-classificationImage ClassificationTransfer Learning

Average-Hard Attention Transformers are Constant-Depth Uniform Threshold Circuits

2023-08-06 · Lena Strobl

Transformers have emerged as a widely used neural network model for various natural language processing tasks. Previous research explored their relationship with constant-depth threshold circuits, making two assumptions:…

Hard Attention