IEEE

Dual Discriminative Low-Rank Projection Learning for Robust Image Classification

Abstract: Numerous methods have exploited projection learning to extract low-dimensional features for image classification. Some projection learning methods integrate low-rank matrix recovery into ...
IEEE

The Orthogonal Projection Component Criterion-Based Order Recursive Least Square for Digital Predistortion of Power Amplifiers

Abstract: In this paper, an orthogonal projection component criterion (OPCC)-based order recursive least square (ORLS) method is proposed for trimming the behavioral models of power amplifiers. The ...
GitHub

ROPIM: Pre-training with Random Orthogonal Projection Image Modeling

[ICLR 2024 Spotlight] The official repository of Self-Supervised Learning methods "ROPIM", Pre-training with Random Orthogonal Projection Image Modeling ROPIM is a self-supervised learning technique ...
GitHub

Πnet: Optimizing hard-constrained neural networks with orthogonal projection layers

This repository contains a JAX implementation of Πnet, an output layer for neural networks that ensures the satisfaction of specified convex constraints. TL;DR: Πnet leverages operator splitting for ...