paper-with-me

홈 › Papers

Inductive Biases and Variable Creation in Self-Attention Mechanisms

2021-10-19 · Benjamin L. Edelman, Surbhi Goel, Sham Kakade, Cyril Zhang

Self-attention, an architectural motif designed to model long-range interactions in sequential data, has driven numerous recent breakthroughs in natural language processing and beyond. This work provides a theoretical analysis of the inductive biases of self-attention modules. Our focus is to rigorously establish which functions and long-range dependencies self-attention blocks prefer to represent. Our main result shows that bounded-norm Transformer networks "create sparse variables": a single self-attention head can represent a sparse function of the input sequence, with sample complexity scaling only logarithmically with the context length. To support our analysis, we present synthetic experiments to probe the sample complexity of learning sparse Boolean functions with Transformers.

📄 PDF Abstract BibTeX arXiv:2110.10090

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
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…
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
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Residual Connection 설명 없음

Similar Papers 제목 키워드 기반

SP-ViT: Learning 2D Spatial Priors for Vision Transformers

2022-06-15 · Yuxuan Zhou, Wangmeng Xiang, Chao Li, Biao Wang 외

Recently, transformers have shown great potential in image classification and established state-of-the-art results on the ImageNet benchmark. However, compared to CNNs, transformers converge slowly and are prone to overf…

image-classificationImage Classification

Towards Flexible Inductive Bias via Progressive Reparameterization Scheduling

2022-10-04 · Yunsung Lee, Gyuseong Lee, Kwangrok Ryoo, Hyojun Go 외

There are two de facto standard architectures in recent computer vision: Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). Strong inductive biases of convolutions help the model learn sample effectivel…

Inductive BiasScheduling

On the Bias Against Inductive Biases

2021-05-28 · George Cazenavette, Simon Lucey

Borrowing from the transformer models that revolutionized the field of natural language processing, self-supervised feature learning for visual tasks has also seen state-of-the-art success using these extremely deep, iso…

ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases

2021-03-19 · Stéphane d'Ascoli, Hugo Touvron, Matthew Leavitt, Ari Morcos 외

Convolutional architectures have proven extremely successful for vision tasks. Their hard inductive biases enable sample-efficient learning, but come at the cost of a potentially lower performance ceiling. Vision Transfo…

image-classificationImage ClassificationInductive Bias

Relational inductive biases on attention mechanisms

2025-07-05 · Víctor Mijangos, Ximena Gutierrez-Vasques, Verónica E. Arriola, Ulises Rodríguez-Domínguez 외 arxiv

Inductive learning aims to construct general models from specific examples, guided by biases that influence hypothesis selection and determine generalization capacity. In this work, we focus on characterizing the relatio…