PyTorch implementation of Rethinking Positional Encoding in Language Pre-training

Related tags

Deep Learningtupe
Overview

TUPE

PyTorch implementation of Rethinking Positional Encoding in Language Pre-training.

alt text

Quickstart

Clone this repository.

git clone https://github.com/jaketae/tupe.git

Navigate to the cloned directory. You can use the bare-bone TUPE Encoder model via

>>> import torch; from tupe import TUPEConfig, TUPEEncoder
>>> config  = TUPEConfig()
>>> model = TUPEEncoder(config)
>>> x = torch.randn(8, 100, 128)
>>> model(x).shape
torch.Size([8, 100, 128])

By default, the model comes with the following parameters:

TUPEConfig(
    num_layers=6, 
    num_heads=8, 
    d_model=128, 
    d_head=16, 
    max_len=256, 
    dropout=0.1, 
    expansion_factor=1, 
    relative_bias=True, 
    bidirectional_bias=True, 
    num_buckets=32, 
    max_distance=128
)

Abstract

In this work, we investigate the positional encoding methods used in language pre- training (e.g., BERT) and identify several problems in the existing formulations. First, we show that in the absolute positional encoding, the addition operation applied on positional embeddings and word embeddings brings mixed correlations between the two heterogeneous information resources. It may bring unnecessary randomness in the attention and further limit the expressiveness of the model. Sec- ond, we question whether treating the position of the symbol [CLS] the same as other words is a reasonable design, considering its special role (the representation of the entire sentence) in the downstream tasks. Motivated from above analysis, we propose a new positional encoding method called Transformer with Untied Positional Encoding (TUPE). In the self-attention module, TUPE computes the word contextual correlation and positional correlation separately with different parameterizations and then adds them together. This design removes the mixed and noisy correlations over heterogeneous embeddings and offers more expres- siveness by using different projection matrices. Furthermore, TUPE unties the [CLS] symbol from other positions, making it easier to capture information from all positions. Extensive experiments and ablation studies on GLUE benchmark demonstrate the effectiveness of the proposed method.

Implementation Notes

  • The default configuration follows TUPE-R, which includes T5's relative position bias. To use TUPE-A, simply toggle TUPEConfig.relative_bias field to False.
  • To avoid limiting the use case of this architecture to BERT-type models with [CLS] tokens, this implementation purposefully omits Section 3.2, on untying the [CLS] symbol from positions.

Citation

@inproceedings{ke2021rethinking,
	title        = {Rethinking Positional Encoding in Language Pre-training},
	author       = {Guolin Ke and Di He and Tie-Yan Liu},
	year         = 2021,
	booktitle    = {International Conference on Learning Representations},
	url          = {https://openreview.net/forum?id=09-528y2Fgf}
}
Owner
Jake Tae
CS + Math @ Yale, SWE intern @huggingface
Jake Tae
Task-based end-to-end model learning in stochastic optimization

Task-based End-to-end Model Learning in Stochastic Optimization This repository is by Priya L. Donti, Brandon Amos, and J. Zico Kolter and contains th

CMU Locus Lab 164 Dec 29, 2022
Implementation for "Conditional entropy minimization principle for learning domain invariant representation features"

Implementation for "Conditional entropy minimization principle for learning domain invariant representation features". The code is reproduced from thi

1 Nov 02, 2022
PyTorch implementation of saliency map-aided GAN for Auto-demosaic+denosing

Saiency Map-aided GAN for RAW2RGB Mapping The PyTorch implementations and guideline for Saiency Map-aided GAN for RAW2RGB Mapping. 1 Implementations B

Yuzhi ZHAO 20 Oct 24, 2022
Random Forests for Regression with Missing Entries

Random Forests for Regression with Missing Entries These are specific codes used in the article: On the Consistency of a Random Forest Algorithm in th

Irving Gómez-Méndez 1 Nov 15, 2021
Studying Python release adoptions by looking at PyPI downloads

Analysis of version adoptions on PyPI We get PyPI download statistics via Google's BigQuery using the pypinfo tool. Usage First you need to get an acc

Julien Palard 9 Nov 04, 2022
Sound and Cost-effective Fuzzing of Stripped Binaries by Incremental and Stochastic Rewriting

StochFuzz: A New Solution for Binary-only Fuzzing StochFuzz is a (probabilistically) sound and cost-effective fuzzing technique for stripped binaries.

Zhuo Zhang 164 Dec 05, 2022
Official implementation of the paper "Steganographer Detection via a Similarity Accumulation Graph Convolutional Network"

SAGCN - Official PyTorch Implementation | Paper | Project Page This is the official implementation of the paper "Steganographer detection via a simila

ZHANG Zhi 1 Nov 26, 2021
Code repository of the paper Neural circuit policies enabling auditable autonomy published in Nature Machine Intelligence

Neural Circuit Policies Enabling Auditable Autonomy Online access via SharedIt Neural Circuit Policies (NCPs) are designed sparse recurrent neural net

8 Jan 07, 2023
Deep GPs built on top of TensorFlow/Keras and GPflow

GPflux Documentation | Tutorials | API reference | Slack What does GPflux do? GPflux is a toolbox dedicated to Deep Gaussian processes (DGP), the hier

Secondmind Labs 107 Nov 02, 2022
Prototype-based Incremental Few-Shot Semantic Segmentation

Prototype-based Incremental Few-Shot Semantic Segmentation Fabio Cermelli, Massimiliano Mancini, Yongqin Xian, Zeynep Akata, Barbara Caputo -- BMVC 20

Fabio Cermelli 21 Dec 29, 2022
A Python type explainer!

typesplainer A Python typehint explainer! Available as a cli, as a website, as a vscode extension, as a vim extension Usage First, install the package

Typesplainer 79 Dec 01, 2022
Collection of machine learning related notebooks to share.

ML_Notebooks Collection of machine learning related notebooks to share. Notebooks GAN_distributed_training.ipynb In this Notebook, TensorFlow's tutori

Sascha Kirch 14 Dec 22, 2022
This repository contains the PyTorch implementation of the paper STaCK: Sentence Ordering with Temporal Commonsense Knowledge appearing at EMNLP 2021.

STaCK: Sentence Ordering with Temporal Commonsense Knowledge This repository contains the pytorch implementation of the paper STaCK: Sentence Ordering

Deep Cognition and Language Research (DeCLaRe) Lab 23 Dec 16, 2022
Wider or Deeper: Revisiting the ResNet Model for Visual Recognition

ademxapp Visual applications by the University of Adelaide In designing our Model A, we did not over-optimize its structure for efficiency unless it w

Zifeng Wu 338 Dec 12, 2022
Hummingbird compiles trained ML models into tensor computation for faster inference.

Hummingbird Introduction Hummingbird is a library for compiling trained traditional ML models into tensor computations. Hummingbird allows users to se

Microsoft 3.1k Dec 30, 2022
Extreme Lightwegith Portrait Segmentation

Extreme Lightwegith Portrait Segmentation Please go to this link to download code Requirements python 3 pytorch = 0.4.1 torchvision==0.2.1 opencv-pyt

HYOJINPARK 59 Dec 16, 2022
Scalable, event-driven, deep-learning-friendly backtesting library

...Minimizing the mean square error on future experience. - Richard S. Sutton BTGym Scalable event-driven RL-friendly backtesting library. Build on

Andrew 922 Dec 27, 2022
Shared Attention for Multi-label Zero-shot Learning

Shared Attention for Multi-label Zero-shot Learning Overview This repository contains the implementation of Shared Attention for Multi-label Zero-shot

dathuynh 26 Dec 14, 2022
Code and data form the paper BERT Got a Date: Introducing Transformers to Temporal Tagging

BERT Got a Date: Introducing Transformers to Temporal Tagging Satya Almasian*, Dennis Aumiller*, and Michael Gertz Heidelberg University Contact us vi

54 Dec 04, 2022
Official repository for ABC-GAN

ABC-GAN The work represented in this repository is the result of a 14 week semesterthesis on photo-realistic image generation using generative adversa

IgorSusmelj 10 Jun 23, 2022