Meta-autoencoders: An approach to discovery and representation of relationships between dynamically evolving classes

TL;DR

This paper introduces meta-autoencoders, neural networks that learn compact representations of collections of autoencoders, enabling the modeling of relationships between dynamically evolving classes, with applications in biological evolution.

Contribution

The paper proposes the concept of meta-autoencoders, extending autoencoders to collections of autoencoders, and explores their potential in modeling evolving classes such as species.

Findings

Initial examples of meta-autoencoders demonstrate their ability to represent class relationships.

Meta-autoencoders can capture properties of evolving classes across different domains.

The approach opens new research directions in machine learning and biology.

Abstract

An autoencoder (AE) is a neural network that, using self-supervised training, learns a succinct parameterized representation, and a corresponding encoding and decoding process, for all instances in a given class. Here, we introduce the concept of a meta-autoencoder (MAE): an AE for a collection of autoencoders. Given a family of classes that differ from each other by the values of some parameters, and a trained AE for each class, an MAE for the family is a neural net that has learned a compact representation and associated encoder and decoder for the class-specific AEs. One application of this general concept is in research and modeling of natural evolution -- capturing the defining and the distinguishing properties across multiple species that are dynamically evolving from each other and from common ancestors. In this interim report we provide a constructive definition of MAEs, initial examples, and the motivating research directions in machine learning and biology.