paper-with-me

홈 › Papers

Investigating the Role of Feed-Forward Networks in Transformers Using Parallel Attention and Feed-Forward Net Design

2023-05-22 · Shashank Sonkar, Richard G. Baraniuk

This paper investigates the key role of Feed-Forward Networks (FFNs) in transformer models by utilizing the Parallel Attention and Feed-Forward Net Design (PAF) architecture, and comparing it to their Series Attention and Feed-Forward Net Design (SAF) counterparts. Central to the effectiveness of PAF are two main assumptions regarding the FFN block and the attention block within a layer: 1) the primary function of the FFN block is to maintain isotropy among token embeddings and prevent their degeneration, and 2) the residual norm computed in the attention block is substantially smaller than the input token embedding norm. To empirically validate these assumptions, we train PAF variants of two large language models (RoBERTa-large and bert-large-uncased). Our results demonstrate that both assumptions hold true in the PAF design. This study contributes to a deeper understanding of the roles and interactions between FFNs and self-attention mechanisms in transformer architectures.

📄 PDF Abstract BibTeX arXiv:2305.13297

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 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 $x$…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
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…
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…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
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

Blockwise Parallel Transformers for Large Context Models

2023-09-21 · NeurIPS 2023 11

Transformers have emerged as the cornerstone of state-of-the-art natural language processing models, showcasing exceptional performance across a wide range of AI applications. However, the memory demands posed by the sel…

Blockwise Parallel Transformer for Large Context Models

2023-05-30 · Hao liu, Pieter Abbeel

Transformers have emerged as the cornerstone of state-of-the-art natural language processing models, showcasing exceptional performance across a wide range of AI applications. However, the memory demands posed by the sel…

Language ModelingLanguage Modelling

Addressing Some Limitations of Transformers with Feedback Memory

2020-02-21 · Angela Fan, Thibaut Lavril, Edouard Grave, Armand Joulin 외

Transformers have been successfully applied to sequential, auto-regressive tasks despite being feedforward networks. Unlike recurrent neural networks, Transformers use attention to capture temporal relations while proces…

Language ModelingLanguage ModellingMachine TranslationReinforcement Learning+1

Universal Transformers

2018-07-10 · ICLR 2019 5 · Mostafa Dehghani, Stephan Gouws, Oriol Vinyals, Jakob Uszkoreit 외

Recurrent neural networks (RNNs) sequentially process data by updating their state with each new data point, and have long been the de facto choice for sequence modeling tasks. However, their inherently sequential comput…

Inductive BiasLAMBADALanguage ModelingLanguage Modelling+3

On the Convergence of Gradient Descent on Learning Transformers with Residual Connections

2025-06-05 · Zhen Qin, Jinxin Zhou, Zhihui Zhu

Transformer models have emerged as fundamental tools across various scientific and engineering disciplines, owing to their outstanding performance in diverse applications. Despite this empirical success, the theoretical …