Convolutional Neural Process

Convolutional Neural Processes (ConvNPs) are a class of neural networks designed to improve data efficiency and uncertainty quantification in prediction tasks, particularly those with limited data. Current research focuses on enhancing ConvNP architectures through techniques like incorporating translation equivariance via convolutional layers, modeling dependencies directly within the prediction process, and developing efficient multi-scale approaches for high-resolution data like images. These advancements are significant because they enable more accurate and robust predictions in various domains, including image restoration, condition monitoring, and time series analysis, while addressing computational limitations associated with traditional methods.

Papers

August 18, 2024

Convolutional Conditional Neural Processes
Wessel P. Bruinsma
Neural Process Conditional Neural Process Convolutional Neural Process Convolutional Conditional Neural Process

March 26, 2024

Serpent: Scalable and Efficient Image Restoration via Multi-scale Structured State Space Models
Mohammad Shahab Sepehri, Zalan Fabian, Mahdi Soltanolkotabi
Multi Scale Image Restoration State Space Model Convolutional Filter Convolutional Neural Process High Resolution Image Restoration

March 25, 2024

Real-time Adaptation for Condition Monitoring Signal Prediction using Label-aware Neural Processes
Seokhyun Chung, Raed Al Kontar
Uncertainty Quantification Condition Monitoring Real Time Adaptation Real Time Prediction Convolutional Neural Process

November 24, 2023

Deep Latent Force Models: ODE-based Process Convolutions for Bayesian Deep Learning
Thomas Baldwin-McDonald, Mauricio A. Álvarez
Uncertainty Quantification Bayesian Deep Learning Deep Gaussian Process Stochastic Variational Inference Convolutional Neural Process Latent Force Model

August 28, 2023

Entropy-based Guidance of Deep Neural Networks for Accelerated Convergence and Improved Performance
Mackenzie J. Meni, Ryan T. White, Michael Mayo, Kevin Pilkiewicz
Neural Network Deep Neural Network High Dimensional Latent Representation Performance Improvement Accelerated Convergence Convolutional Neural Process Robust Entropy Adaptive Loss Minimization

February 26, 2023

Path Integral Based Convolution and Pooling for Heterogeneous Graph Neural Networks
Lingjie Kong, Yun Liao
Convolutional Neural Network Heterogeneous Graph Neural Network Graph Prediction Convolutional Neural Process

September 30, 2022

Blind Signal Dereverberation for Machine Speech Recognition
Samik Sadhu, Hynek Hermansky
Automatic Speech Recognition Speech Dereverberation Reverberant Environment Noisy Reverberant Convolutional Neural Process

June 17, 2022

Shallow and Deep Nonparametric Convolutions for Gaussian Processes
Thomas M. McDonald, Magnus Ross, Michael T. Smith, Mauricio A. Álvarez
Gaussian Process Shallow Water Covariance Function Convolutional Neural Process

May 24, 2022

Convolutional Neural Processes for Inpainting Satellite Images
Alexander Pondaven, Märt Bakler, Donghu Guo, Hamzah Hashim, Martin Ignatov, Harrison Zhu
Satellite Image Distribution Image Convolutional Neural Process

March 24, 2022

Rubik's Cube Operator: A Plug And Play Permutation Module for Better Arranging High Dimensional Industrial Data in Deep Convolutional Processes
Luoxiao Yang, Zhong Zheng, Zijun Zhang
Convolutional Neural Network High Dimensional Data Wind Power Feature Operator Machine Permutation Industrial Data Input Tensor Convolutional Neural Process

Convolutional Neural Process

Papers

Convolutional Conditional Neural Processes

Serpent: Scalable and Efficient Image Restoration via Multi-scale Structured State Space Models

Real-time Adaptation for Condition Monitoring Signal Prediction using Label-aware Neural Processes

Deep Latent Force Models: ODE-based Process Convolutions for Bayesian Deep Learning

Entropy-based Guidance of Deep Neural Networks for Accelerated Convergence and Improved Performance

Path Integral Based Convolution and Pooling for Heterogeneous Graph Neural Networks

Blind Signal Dereverberation for Machine Speech Recognition

Shallow and Deep Nonparametric Convolutions for Gaussian Processes

Convolutional Neural Processes for Inpainting Satellite Images

Rubik's Cube Operator: A Plug And Play Permutation Module for Better Arranging High Dimensional Industrial Data in Deep Convolutional Processes