Show simple item record

AuthorMalik J.
AuthorKiranyaz, Mustafa Serkan
AuthorGabbouj M.
Available date2022-04-26T12:31:18Z
Publication Date2021
Publication NameNeural Networks
ResourceScopus
Identifierhttp://dx.doi.org/10.1016/j.neunet.2020.12.014
URIhttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85098704465&doi=10.1016%2fj.neunet.2020.12.014&partnerID=40&md5=b34a472199addec3947e6c9bfaaf1dba
URIhttp://hdl.handle.net/10576/30591
AbstractDiscriminative learning based on convolutional neural networks (CNNs) aims to perform image restoration by learning from training examples of noisy-clean image pairs. It has become the go-to methodology for tackling image restoration and has outperformed the traditional non-local class of methods. However, the top-performing networks are generally composed of many convolutional layers and hundreds of neurons, with trainable parameters in excess of several million. We claim that this is due to the inherently linear nature of convolution-based transformation, which is inadequate for handling severe restoration problems. Recently, a non-linear generalization of CNNs, called the operational neural networks (ONN), has been shown to outperform CNN on AWGN denoising. However, its formulation is burdened by a fixed collection of well-known non-linear operators and an exhaustive search to find the best possible configuration for a given architecture, whose efficacy is further limited by a fixed output layer operator assignment. In this study, we leverage the Taylor series-based function approximation to propose a self-organizing variant of ONNs, Self-ONNs, for image restoration, which synthesizes novel nodal transformations on-the-fly as part of the learning process, thus eliminating the need for redundant training runs for operator search. In addition, it enables a finer level of operator heterogeneity by diversifying individual connections of the receptive fields and weights. We perform a series of extensive ablation experiments across three severe image restoration tasks. Even when a strict equivalence of learnable parameters is imposed, Self-ONNs surpass CNNs by a considerable margin across all problems, improving the generalization performance by up to 3 dB in terms of PSNR.
Languageen
PublisherElsevier Ltd
SubjectConvolution
Convolutional neural networks
Linear transformations
Mathematical operators
Mathematical transformations
Personnel training
Restoration
Ablation experiments
Discriminative learning
Function approximation
Generalization performance
Image restoration problem
Receptive fields
Restoration problems
Strict equivalence
Image reconstruction
article
convolutional neural network
drug efficacy
human
image reconstruction
learning
receptive field
automated pattern recognition
image processing
nerve cell
photostimulation
physiology
procedures
Humans
Image Processing, Computer-Assisted
Neural Networks, Computer
Neurons
Pattern Recognition, Automated
Photic Stimulation
TitleSelf-organized operational neural networks for severe image restoration problems
TypeArticle
Pagination201-211
Volume Number135
dc.accessType Abstract Only


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record