paper-with-me

Papers

BETA: Binarized Energy-Efficient Transformer Accelerator at the Edge

2024-01-22 · Yuhao Ji, Chao Fang, Zhongfeng Wang

Existing binary Transformers are promising in edge deployment due to their compact model size, low computational complexity, and considerable inference accuracy. However, deploying binary Transformers faces challenges on prior processors due to inefficient execution of quantized matrix multiplication (QMM) and the energy consumption overhead caused by multi-precision activations. To tackle the challenges above, we first develop a computation flow abstraction method for binary Transformers to improve QMM execution efficiency by optimizing the computation order. Furthermore, a binarized energy-efficient Transformer accelerator, namely BETA, is proposed to boost the efficient deployment at the edge. Notably, BETA features a configurable QMM engine, accommodating diverse activation precisions of binary Transformers and offering high-parallelism and high-speed for QMMs with impressive energy efficiency. Experimental results evaluated on ZCU102 FPGA show BETA achieves an average energy efficiency of 174 GOPS/W, which is 1.76~21.92x higher than prior FPGA-based accelerators, showing BETA's good potential for edge Transformer acceleration.

📄 PDF Abstract BibTeX arXiv:2401.11851

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
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…
Residual Connection 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…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Similar Papers 제목 키워드 기반

Co-Designing Binarized Transformer and Hardware Accelerator for Efficient End-to-End Edge Deployment

2024-07-16 · Yuhao Ji, Chao Fang, Shaobo Ma, Haikuo Shao 외

Transformer models have revolutionized AI tasks, but their large size hinders real-world deployment on resource-constrained and latency-critical edge devices. While binarized Transformers offer a promising solution by si…

QuantizationScheduling

A Fully Hardware Implemented Accelerator Design in ReRAM Analog Computing without ADCs

2024-12-27 · Peng Dang, Huawei Li, Wei Wang

Emerging ReRAM-based accelerators process neural networks via analog Computing-in-Memory (CiM) for ultra-high energy efficiency. However, significant overhead in peripheral circuits and complex nonlinear activation modes…

ROBIN: A Robust Optical Binary Neural Network Accelerator

2021-07-12 · Febin P. Sunny, Asif Mirza, Mahdi Nikdast, Sudeep Pasricha

Domain specific neural network accelerators have garnered attention because of their improved energy efficiency and inference performance compared to CPUs and GPUs. Such accelerators are thus well suited for resource-con…

Binarized Convolutional Neural Networks with Separable Filters for Efficient Hardware Acceleration

2017-07-15 · Jeng-Hau Lin, Tianwei Xing, Ritchie Zhao, Zhiru Zhang 외

State-of-the-art convolutional neural networks are enormously costly in both compute and memory, demanding massively parallel GPUs for execution. Such networks strain the computational capabilities and energy available t…

COBRA: Algorithm-Architecture Co-optimized Binary Transformer Accelerator for Edge Inference

2025-04-22 · Ye Qiao, Zhiheng Chen, Yian Wang, Yifan Zhang 외

Transformer-based models have demonstrated superior performance in various fields, including natural language processing and computer vision. However, their enormous model size and high demands in computation, memory, an…

Edge-computing