Dissipative residual layers for unsupervised implicit parameterization of data manifolds

10/13/2022
by   Viktor Reshniak, et al.
0

We propose an unsupervised technique for implicit parameterization of data manifolds. In our approach, the data is assumed to belong to a lower dimensional manifold in a higher dimensional space, and the data points are viewed as the endpoints of the trajectories originating outside the manifold. Under this assumption, the data manifold is an attractive manifold of a dynamical system to be estimated. We parameterize such a dynamical system with a residual neural network and propose a spectral localization technique to ensure it is locally attractive in the vicinity of data. We also present initialization and additional regularization of the proposed residual layers. that we call dissipative bottlenecks. We mention the importance of the considered problem for the tasks of reinforcement learning and support our discussion with examples demonstrating the performance of the proposed layers in denoising and generative tasks.

READ FULL TEXT

page 4

page 5

research
06/22/2022

Neural Implicit Manifold Learning for Topology-Aware Generative Modelling

Natural data observed in ℝ^n is often constrained to an m-dimensional ma...
research
10/14/2021

Inferring Manifolds From Noisy Data Using Gaussian Processes

In analyzing complex datasets, it is often of interest to infer lower di...
research
08/21/2019

A Neural Network for Semi-Supervised Learning on Manifolds

Semi-supervised learning algorithms typically construct a weighted graph...
research
06/01/2018

Intrinsic Isometric Manifold Learning with Application to Localization

Data living on manifolds commonly appear in many applications. We show t...
research
10/30/2020

The Mathematical Foundations of Manifold Learning

Manifold learning is a popular and quickly-growing subfield of machine l...
research
05/01/2023

Autoencoders for discovering manifold dimension and coordinates in data from complex dynamical systems

While many phenomena in physics and engineering are formally high-dimens...
research
09/17/2020

LAAT: Locally Aligned Ant Technique for detecting manifolds of varying density

Dimensionality reduction and clustering are often used as preliminary st...

Please sign up or login with your details

Forgot password? Click here to reset