On Complex Valued Convolutional Neural Networks

Abstract

Convolutional neural networks (CNNs) are the cutting edge model for supervised machine learning in computer vision. In recent years CNNs have outperformed traditional approaches in many computer vision tasks such as object detection, image classification and face recognition. CNNs are vulnerable to overfitting, and a lot of research focuses on finding regularization methods to overcome it. One approach is designing task specific models based on prior knowledge. Several works have shown that properties of natural images can be easily captured using complex numbers. Motivated by these works, we present a variation of the CNN model with complex valued input and weights. We construct the complex model as a generalization of the real model. Lack of order over the complex field raises several difficulties both in the definition and in the training of the network. We address these issues and suggest possible solutions. The resulting model is shown to be a restricted form of a real valued CNN with twice the parameters. It is sensitive to phase structure, and we suggest it serves as a regularized model for problems where such structure is important. This suggestion is verified empirically by comparing the performance of a complex and a real network in the problem of cell detection. The two networks achieve comparable results, and although the complex model is hard to train, it is significantly less vulnerable to overfitting. We also demonstrate that the complex network detects meaningful phase structure in the data.

M.Sc. thesis

References in corpus (3)

• Deep Learning in Neural Networks: An Overview
• A mathematical motivation for complex-valued convolutional networks
• Complex-Valued Hough Transforms for Circles

Cited by in corpus (17)

• Deep Complex Networks
• Prior-Guided Deep Interference Mitigation for FMCW Radars
• Deep Complex Networks for Protocol-Agnostic Radio Frequency Device Fingerprinting in the Wild
• Building Blocks for a Complex-Valued Transformer Architecture
• Evaluation of Complex-Valued Neural Networks on Real-Valued Classification Tasks
• 3D-Rotation-Equivariant Quaternion Neural Networks
• Lending Orientation to Neural Networks for Cross-view Geo-localization
• Co-VeGAN: Complex-Valued Generative Adversarial Network for Compressive Sensing MR Image Reconstruction
• Complex-valued Federated Learning with Differential Privacy and MRI Applications
• Image classification in frequency domain with 2SReLU: a second harmonics superposition activation function
• Semi-supervised Complex-valued GAN for Polarimetric SAR Image Classification
• Neural Networks with Complex-Valued Weights Have No Spurious Local Minima
• Theoretical Exploration of Flexible Transmitter Model
• Widely Linear Kernels for Complex-Valued Kernel Activation Functions
• Bayesian Sparsification Methods for Deep Complex-valued Networks
• Complex Amplitude-Phase Boltzmann Machines
• Implementation of Optical Deep Neural Networks using the Fabry-Perot Interferometer