Page Not Found
Page not found. Your pixels are in another canvas.
Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
About
About me
Page not in menu
This is a page not in th emain menu
Posts
portfolio
publications
Weight Compander: A Simple Weight Reparameterization for Regularization
Published in IJCNN, 2023 [ArXiv]
Regularization is a set of techniques that are used to improve the generalization ability of deep neural networks. In this paper, we introduce weight compander (WC), a novel effective method to improve generalization by reparameterizing each weight in deep neural networks using a nonlinear function.
Spectral Batch Normalization: Normalization in the Frequency Domain
Published in IJCNN, 2023 [ArXiv]
Regularization is a set of techniques that are used to improve the generalization ability of deep neural networks. In this paper, we introduce spectral batch normalization (SBN), a novel effective method to improve generalization by normalizing feature maps in the frequency (spectral) domain.
Spectral Wavelet Dropout: Regularization in the Wavelet Domain
Published in ICMLA, 2024 [ArXiv]
This work introduces Spectral Wavelet Dropout (SWD), a novel regularization method that includes two variants: 1D-SWD and 2D-SWD. These variants improve CNN generalization by randomly dropping detailed frequency bands in the discrete wavelet decomposition of feature maps.
Squeeze-and-Remember Block
Published in ICMLA, 2024 [ArXiv]
CNNs lack a dynamic feature retention mechanism similar to the human brain’s memory, limiting their ability to use learned information in new contexts. To bridge this gap, we introduce the “Squeeze-and-Remember” (SR) block, a novel architectural unit that gives CNNs dynamic memory-like functionalities.
CNN Mixture-of-Depths
Published in ACCV, 2024 [ArXiv]
We introduce Mixture-of-Depths (MoD) for Convolutional Neural Networks (CNNs), a novel approach that enhances the computational efficiency of CNNs by selectively processing channels based on their relevance to the current prediction.