Transfer Learning for Dead Fuel Moisture Prediction Using Time-Warping Recurrent Neural Networks

TL;DR

This paper introduces a time-warping transfer learning approach using LSTM-based RNNs to adapt fuel moisture prediction models across different fuel classes with limited data.

Contribution

It presents a novel time-warping transfer learning method that enables RNNs to transfer learned dynamics across fuel moisture classes with sparse data.

Findings

The method successfully transfers predictions across fuel classes.

It improves prediction accuracy for underrepresented fuel classes.

Validation shows comparable or better performance than existing models.

Abstract

This paper proposes a time-warping transfer learning method, a technique for temporally rescaling the learned dynamics of a recurrent neural network (RNN) with a Long Short-Term Memory (LSTM) layer to enable task transfer across fuel moisture classes. Fuel moisture content (FMC) is divided into idealized classes based on characteristic lag time. Large quantities of real-time data are available for 10h fuels from sensors on weather stations, but observations of other fuel classes are sparse in space and time. We use transfer learning to adapt an RNN pretrained on 10h FMC to predict FMC for 1h, 100h, and 1000h fuels. We validate this method using data from a landmark field study conducted in Oklahoma that was used to calibrate the state-of-the-art Nelson fuel moisture model.