# Multi-Model Least Squares-Based Recomputation Framework for Large Data Analysis

Source: [Multi-Model Least Squares-Based Recomputation Framework for Large Data Analysis](<https://www.sota2.com/research/paper/68fa69751e209131cea22398>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T23:00:47.140Z

Authors: Wandong Zhang, QM Jonathan Wu, Yimin Yang, WG Will Zhao, Tianlei Wang, Hui Zhang (6) \_\_INSTITUTION\_6\_\_ University of Windsor, (2) Lakehead University, (3) Vector Institute for Artificial Intelligence, (4) CEGEP de Ste Foy, (5) Hangzhou Dianzi University, (6) Hunan University)

Published: 2021-01-04T23:01:30.000Z

## Abstract

Most multilayer least squares (LS)-based neural networks are structured with two separate stages: unsupervised feature encoding and supervised pattern classification. Once the unsupervised learning is finished, the latent encoding would be fixed without supervised fine-tuning. However, in complex tasks such as handling the ImageNet dataset, there are often many more clues that can be directly encoded, while the unsupervised learning, by definition cannot know exactly what is useful for a certain task. This serves as the motivation to retrain the latent space representations to learn some clues that unsupervised learning has not yet learned. In particular, the error matrix from the output layer is pulled back to each hidden layer, and the parameters of the hidden layer are recalculated with Moore-Penrose (MP) inverse for more generalized representations. In this paper, a recomputation-based multilayer network using MP inverse (RML-MP) is developed. A sparse RML-MP (SRML-MP) model to boost the performance of RML-MP is then proposed. The experimental results with varying training samples (from 3 K to 1.8 M) show that the proposed models provide better generalization performance than most representation learning algorithms.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2101.01271>)

[PDF](<https://arxiv.org/pdf/2101.01271.pdf>)
