Loading Now

Summary of Inductive Global and Local Manifold Approximation and Projection, by Jungeum Kim and Xiao Wang


Inductive Global and Local Manifold Approximation and Projection

by Jungeum Kim, Xiao Wang

First submitted to arxiv on: 12 Jun 2024

Categories

  • Main: Machine Learning (cs.LG)
  • Secondary: Applications (stat.AP); Methodology (stat.ME)

     Abstract of paper      PDF of paper


GrooveSquid.com Paper Summaries

GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!

Summary difficulty Written by Summary
High Paper authors High Difficulty Summary
Read the original abstract here
Medium GrooveSquid.com (original content) Medium Difficulty Summary
The paper proposes a novel manifold learning method called GLoMAP for nonlinear dimensional reduction and high-dimensional data visualization. Building upon the success of t-SNE and UMAP, this method preserves local and global structure information in the data. A unique aspect is its ability to display a progression from global to local formation during optimization. The authors also introduce an inductive version, iGLoMAP, which uses a deep neural network to generate lower-dimensional embeddings for unseen points without re-training. This enables mini-batch learning for large-scale accelerated gradient calculations. Experiments demonstrate competitive results against state-of-the-art methods on both simulated and real-world datasets.
Low GrooveSquid.com (original content) Low Difficulty Summary
This paper creates new ways to shrink big data into smaller, more understandable pieces. It uses a special kind of math called manifold learning to help us understand patterns in the data. The method is good at showing both small details and big pictures. The authors also created an improved version that can work with new data without needing to re-do everything. They tested it on real-world datasets and compared it to other methods, finding similar results.

Keywords

» Artificial intelligence  » Manifold learning  » Neural network  » Optimization  » Umap