paper-with-me

홈 › Papers

Parameter-Efficient Transfer Learning for NLP

2019-02-02 · Neil Houlsby, Andrei Giurgiu, Stanislaw Jastrzebski, Bruna Morrone, Quentin de Laroussilhe, Andrea Gesmundo, Mona Attariyan, Sylvain Gelly

Fine-tuning large pre-trained models is an effective transfer mechanism in NLP. However, in the presence of many downstream tasks, fine-tuning is parameter inefficient: an entire new model is required for every task. As an alternative, we propose transfer with adapter modules. Adapter modules yield a compact and extensible model; they add only a few trainable parameters per task, and new tasks can be added without revisiting previous ones. The parameters of the original network remain fixed, yielding a high degree of parameter sharing. To demonstrate adapter's effectiveness, we transfer the recently proposed BERT Transformer model to 26 diverse text classification tasks, including the GLUE benchmark. Adapters attain near state-of-the-art performance, whilst adding only a few parameters per task. On GLUE, we attain within 0.4% of the performance of full fine-tuning, adding only 3.6% parameters per task. By contrast, fine-tuning trains 100% of the parameters per task.

📄 PDF Abstract BibTeX arXiv:1902.00751

Code (17)

google-research/adapter-bert 공식 구현 tf
Adapter-Hub/adapter-transformers pytorch
AsaCooperStickland/Bert-n-Pals pytorch
CyberZHG/keras-bert tf
Davidzhangyuanhan/NOAH tf
TATlong/keras_bert tf
ZhangYuanhan-AI/NOAH tf
cs-mshah/Adapter-Bert pytorch
heekhero/DTL pytorch
hmohebbi/TF-Adapter-BERT tf
kpe/bert-for-tf2 tf
krypticmouse/Adapter-BERT pytorch
mindspore-courses/d2l-mindspore/tree/master/chapter_15_natural_language_processing_applications mindspore
osu-mlb/petl_vision pytorch
osu-mlb/vit_peft_vision pytorch
prrao87/fine-grained-sentiment pytorch
zphang/bert_on_stilts pytorch

Tasks

Image ClassificationText ClassificationTransfer Learning

Methods 이 논문이 사용한 방법론

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
Residual Connection 설명 없음
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Weight Decay 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…

Similar Papers 제목 키워드 기반

Parameter Transfer Extreme Learning Machine based on Projective Model

2018-09-04 · Chao Chen, Boyuan Jiang, Xinyu Jin

Recent years, transfer learning has attracted much attention in the community of machine learning. In this paper, we mainly focus on the tasks of parameter transfer under the framework of extreme learning machine (ELM). …

Domain Adaptationfeature selectionTransfer Learning

Double Double Descent: On Generalization Errors in Transfer Learning between Linear Regression Tasks

2020-06-12 · Yehuda Dar, Richard G. Baraniuk

We study the transfer learning process between two linear regression problems. An important and timely special case is when the regressors are overparameterized and perfectly interpolate their training data. We examine a…

regressionTransfer Learning

Learning Bound for Parameter Transfer Learning

2016-10-27 · NeurIPS 2016 12 · Wataru Kumagai

We consider a transfer-learning problem by using the parameter transfer approach, where a suitable parameter of feature mapping is learned through one task and applied to another objective task. Then, we introduce the no…

Transfer Learning

Parameter Transfer Unit for Deep Neural Networks

2018-04-23 · Yinghua Zhang, Yu Zhang, Qiang Yang

Parameters in deep neural networks which are trained on large-scale databases can generalize across multiple domains, which is referred as "transferability". Unfortunately, the transferability is usually defined as discr…

Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration

2025-12-26 · Bruno Mlodozeniec, Pierre Ablin, Louis Béthune, Dan Busbridge 외 arxiv

Hyperparameter tuning can dramatically impact training stability and final performance of large-scale models. Recent works on neural network parameterisations, such as $μ$P, have enabled transfer of optimal global hyperp…