# Investigating Recurrent Neural Network Memory Structures using Neuro-Evolution

## TL;DR

This paper introduces EXAMM, a neuro-evolution algorithm that evolves RNNs with various memory structures to predict large-scale real-world time series data, providing insights into memory cell design and evolution.

## Contribution

The paper presents a novel neuro-evolution method, EXAMM, capable of evolving diverse RNN memory structures for complex time series prediction tasks.

## Key findings

- Evolved RNNs achieved high prediction accuracy on real-world datasets.
- Different memory cell types showed varying effectiveness in prediction tasks.
- Statistical analysis of evolved RNNs offers insights for future memory cell design.

## Abstract

This paper presents a new algorithm, Evolutionary eXploration of Augmenting Memory Models (EXAMM), which is capable of evolving recurrent neural networks (RNNs) using a wide variety of memory structures, such as Delta-RNN, GRU, LSTM, MGU and UGRNN cells. EXAMM evolved RNNs to perform prediction of large-scale, real world time series data from the aviation and power industries. These data sets consist of very long time series (thousands of readings), each with a large number of potentially correlated and dependent parameters. Four different parameters were selected for prediction and EXAMM runs were performed using each memory cell type alone, each cell type with feed forward nodes, and with all possible memory cell types. Evolved RNN performance was measured using repeated k-fold cross validation, resulting in 1210 EXAMM runs which evolved 2,420,000 RNNs in 12,100 CPU hours on a high performance computing cluster. Generalization of the evolved RNNs was examined statistically, providing interesting findings that can help refine the RNN memory cell design as well as inform future neuro-evolution algorithms development.