# USEFUSE: Uniform Stride for Enhanced Performance in Fused Layer Architecture of Deep Neural Networks

arXiv: https://arxiv.org/abs/2412.13724 · HTML: https://mlsanity.com/paper/2412.13724/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 65 references that are in the map.
- First posted on arXiv 18 Dec 2024.

## Builds on (most cited first, 20 of 20)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Plug-and-Play Image Restoration with Deep Denoiser Prior](https://mlsanity.com/paper/2008.13751/) (2020)
- [Bit Fusion: Bit-Level Dynamically Composable Architecture for Accelerating Deep Neural Networks](https://mlsanity.com/paper/1712.01507/) (2017)
- [Practical Blind Image Denoising via Swin-Conv-UNet and Data Synthesis](https://mlsanity.com/paper/2203.13278/) (2022)
- [A Study on ReLU and Softmax in Transformer](https://mlsanity.com/paper/2302.06461/) (2023)
- [Automatically designing CNN architectures using genetic algorithm for image classification](https://mlsanity.com/paper/1808.03818/) (2018)
- [Replacing softmax with ReLU in Vision Transformers](https://mlsanity.com/paper/2309.08586/) (2023)
- [ArtAdapter: Text-to-Image Style Transfer using Multi-Level Style Encoder and Explicit Adaptation](https://mlsanity.com/paper/2312.02109/) (2023)
- [Bit-pragmatic Deep Neural Network Computing](https://mlsanity.com/paper/1610.06920/) (2016)
- [AdaPT: Fast Emulation of Approximate DNN Accelerators in PyTorch](https://mlsanity.com/paper/2203.04071/) (2022)
- [Efficient B-mode Ultrasound Image Reconstruction from Sub-sampled RF Data using Deep Learning](https://mlsanity.com/paper/1712.06096/) (2017)
- [Fixed-point optimization of deep neural networks with adaptive step size retraining](https://mlsanity.com/paper/1702.08171/) (2017)
- [ShortcutFusion: From Tensorflow to FPGA-based accelerator with reuse-aware memory allocation for shortcut data](https://mlsanity.com/paper/2106.08167/) (2021)
- [Design of High-Throughput Mixed-Precision CNN Accelerators on FPGA](https://mlsanity.com/paper/2208.04854/) (2022)
- [Low-Latency Online Multiplier with Reduced Activities and Minimized Interconnect for Inner Product Arrays](https://mlsanity.com/paper/2304.12946/) (2023)
- [CompRRAE: RRAM-based Convolutional Neural Network Accelerator with Reduced Computations through a Runtime Activation Estimation](https://mlsanity.com/paper/1906.03180/) (2019)
- [AFP-SRC: Identification of Antifreeze Proteins Using Sparse Representation Classifier](https://mlsanity.com/paper/2009.05277/) (2020)
- [CITISEN: A Deep Learning-Based Speech Signal-Processing Mobile Application](https://mlsanity.com/paper/2008.09264/) (2020)

## Built on by (most cited first, 1 of 1)

- [Energy-Efficient CNN Acceleration with MSDF Digit-Serial Arithmetic on FPGA](https://mlsanity.com/paper/2606.25562/) (2026)

## Abstract (arXiv)

Convolutional Neural Networks (CNNs) are crucial in various applications, but their deployment on resource-constrained edge devices poses challenges. This study presents the Sum-of-Products (SOP) units for convolution, which utilize low-latency left-to-right bit-serial arithmetic to minimize response time and enhance overall performance. The study proposes a methodology for fusing multiple convolution layers to reduce off-chip memory communication and increase overall performance. An effective mechanism detects and skips inefficient convolutions after ReLU layers, minimizing power consumption without compromising accuracy. Furthermore, efficient tile movement guarantees uniform access to the fusion pyramid. An analysis demonstrates the utile stride strategy improves operational intensity. Two designs cater to varied demands: one focuses on minimal response time for mission-critical applications, and another focuses on resource-constrained devices with comparable latency. This approach notably reduced redundant computations, improving the efficiency of CNN deployment on edge devices.