# Kernel-Segregated Transpose Convolution Operation

arXiv: https://arxiv.org/abs/2209.03704 · HTML: https://mlsanity.com/paper/2209.03704/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 28 references that are in the map.
- First posted on arXiv 8 Sep 2022.

## Builds on (most cited first, 9 of 9)

- [Caffe: Convolutional Architecture for Fast Feature Embedding](https://mlsanity.com/paper/1408.5093/) (2014)
- [Fast Algorithms for Convolutional Neural Networks](https://mlsanity.com/paper/1509.09308/) (2015)
- [TVM: An Automated End-to-End Optimizing Compiler for Deep Learning](https://mlsanity.com/paper/1802.04799/) (2018)
- [Fast Convolutional Nets With fbfft: A GPU Performance Evaluation](https://mlsanity.com/paper/1412.7580/) (2014)
- [High Performance Zero-Memory Overhead Direct Convolutions](https://mlsanity.com/paper/1809.10170/) (2018)
- [Anatomy Of High-Performance Deep Learning Convolutions On SIMD Architectures](https://mlsanity.com/paper/1808.05567/) (2018)
- [GANAX: A Unified MIMD-SIMD Acceleration for Generative Adversarial Networks](https://mlsanity.com/paper/1806.01107/) (2018)
- [An Energy-Efficient FPGA-based Deconvolutional Neural Networks Accelerator for Single Image Super-Resolution](https://mlsanity.com/paper/1801.05997/) (2018)
- [The Indirect Convolution Algorithm](https://mlsanity.com/paper/1907.02129/) (2019)

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

- [Unified Kernel-Segregated Transpose Convolution Operation](https://mlsanity.com/paper/2502.20493/) (2025)

## Abstract (arXiv)

Transpose convolution has shown prominence in many deep learning applications. However, transpose convolution layers are computationally intensive due to the increased feature map size due to adding zeros after each element in each row and column. Thus, convolution operation on the expanded input feature map leads to poor utilization of hardware resources. The main reason for unnecessary multiplication operations is zeros at predefined positions in the input feature map. We propose an algorithmic-level optimization technique for the effective transpose convolution implementation to solve these problems. Based on kernel activations, we segregated the original kernel into four sub-kernels. This scheme could reduce memory requirements and unnecessary multiplications. Our proposed method was 3.09 (3.02) × faster computation using the Titan X GPU (Intel Dual Core CPU) with a flower dataset from the Kaggle website. Furthermore, the proposed optimization method can be generalized to existing devices without additional hardware requirements. A simple deep learning model containing one transpose convolution layer was used to evaluate the optimization method. It showed 2.2 × faster training using the MNIST dataset with an Intel Dual-core CPU than the conventional implementation.