English

IAE-Net: Integral Autoencoders for Discretization-Invariant Learning

Machine Learning 2022-09-07 v3 Numerical Analysis Numerical Analysis

Abstract

Discretization invariant learning aims at learning in the infinite-dimensional function spaces with the capacity to process heterogeneous discrete representations of functions as inputs and/or outputs of a learning model. This paper proposes a novel deep learning framework based on integral autoencoders (IAE-Net) for discretization invariant learning. The basic building block of IAE-Net consists of an encoder and a decoder as integral transforms with data-driven kernels, and a fully connected neural network between the encoder and decoder. This basic building block is applied in parallel in a wide multi-channel structure, which are repeatedly composed to form a deep and densely connected neural network with skip connections as IAE-Net. IAE-Net is trained with randomized data augmentation that generates training data with heterogeneous structures to facilitate the performance of discretization invariant learning. The proposed IAE-Net is tested with various applications in predictive data science, solving forward and inverse problems in scientific computing, and signal/image processing. Compared with alternatives in the literature, IAE-Net achieves state-of-the-art performance in existing applications and creates a wide range of new applications.

Keywords

Cite

@article{arxiv.2203.05142,
  title  = {IAE-Net: Integral Autoencoders for Discretization-Invariant Learning},
  author = {Yong Zheng Ong and Zuowei Shen and Haizhao Yang},
  journal= {arXiv preprint arXiv:2203.05142},
  year   = {2022}
}

Comments

submitted to JMLR. revised version following review results

R2 v1 2026-06-24T10:08:10.724Z