paper-with-me

Papers

Shatter: An Efficient Transformer Encoder with Single-Headed Self-Attention and Relative Sequence Partitioning

2021-08-30 · Ran Tian, Joshua Maynez, Ankur P. Parikh

The highly popular Transformer architecture, based on self-attention, is the foundation of large pretrained models such as BERT, that have become an enduring paradigm in NLP. While powerful, the computational resources and time required to pretrain such models can be prohibitive. In this work, we present an alternative self-attention architecture, Shatter, that more efficiently encodes sequence information by softly partitioning the space of relative positions and applying different value matrices to different parts of the sequence. This mechanism further allows us to simplify the multi-headed attention in Transformer to single-headed. We conduct extensive experiments showing that Shatter achieves better performance than BERT, with pretraining being faster per step (15% on TPU), converging in fewer steps, and offering considerable memory savings (>50%). Put together, Shatter can be pretrained on 8 V100 GPUs in 7 days, and match the performance of BERT_Base -- making the cost of pretraining much more affordable.

📄 PDF Abstract BibTeX arXiv:2108.13032

Code (0)

등록된 구현이 없습니다.

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
Weight Decay 설명 없음
Adam 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…
WordPiece 설명 없음

Similar Papers 제목 키워드 기반

Hard-Coded Gaussian Attention for Neural Machine Translation

2020-05-02 · ACL 2020 6 · Weiqiu You, Simeng Sun, Mohit Iyyer

Recent work has questioned the importance of the Transformer's multi-headed attention for achieving high translation quality. We push further in this direction by developing a "hard-coded" attention variant without any l…

DecoderMachine TranslationTranslation

Two-Headed Monster And Crossed Co-Attention Networks

2019-11-10 · Asian Chapter of the Association for Computational Linguistics 2020 · Yaoyiran Li, Jing Jiang

This paper presents some preliminary investigations of a new co-attention mechanism in neural transduction models. We propose a paradigm, termed Two-Headed Monster (THM), which consists of two symmetric encoder modules a…

DecoderTranslationVocal Bursts Valence Prediction

EEG-ConvTransformer for Single-Trial EEG based Visual Stimuli Classification

2021-07-08 · Subhranil Bagchi, Deepti R. Bathula

Different categories of visual stimuli activate different responses in the human brain. These signals can be captured with EEG for utilization in applications such as Brain-Computer Interface (BCI). However, accurate cla…

Brain Computer InterfaceClassificationDiversityEEG+1

Cross-stitched Multi-modal Encoders

2022-01-16 · ACL ARR January 2022 1 · Anonymous

In this paper, we propose a novel architecture for multi-modal speech and text input. We combine pretrained speech and text encoders using multi-headed cross-modal attention and jointly fine-tune on the target problem. T…

GPU

Cross-stitched Multi-modal Encoders

2022-04-20 · Karan Singla, Daniel Pressel, Ryan Price, Bhargav Srinivas Chinnari 외

In this paper, we propose a novel architecture for multi-modal speech and text input. We combine pretrained speech and text encoders using multi-headed cross-modal attention and jointly fine-tune on the target problem. T…

ClassificationGPU