Research on Patch Attentive Neural Process

TL;DR

This paper introduces Patch Attentive Neural Process (PANP), a model that enhances image feature extraction and reconstruction efficiency by using image patches, inspired by ViT and MAE, addressing ANP's time complexity issues.

Contribution

The paper proposes PANP, a novel model that improves upon ANP by utilizing image patches and refined deterministic paths for better accuracy and efficiency.

Findings

Enhanced image feature extraction and reconstruction.

Improved prediction accuracy over traditional ANP.

Reduced time complexity for processing input sequences.

Abstract

Attentive Neural Process (ANP) improves the fitting ability of Neural Process (NP) and improves its prediction accuracy, but the higher time complexity of the model imposes a limitation on the length of the input sequence. Inspired by models such as Vision Transformer (ViT) and Masked Auto-Encoder (MAE), we propose Patch Attentive Neural Process (PANP) using image patches as input and improve the structure of deterministic paths based on ANP, which allows the model to extract image features more accurately and efficiently reconstruction.