Static Seeding and Clustering of LSTM Embeddings to Learn from Loosely Time-Decoupled Events

TL;DR

This paper enhances LSTM-based time series prediction for loosely coupled events by seeding models with static socio-economic embeddings, improving COVID-19 case forecasts at the US county level.

Contribution

It introduces a novel seeding approach using static embeddings to improve LSTM predictions for loosely time-decoupled events like disease spread.

Findings

Improved 10-day COVID-19 case prediction accuracy.

Clustering embeddings identifies similar event patterns.

Seeding with socio-economic data enhances model performance.

Abstract

Humans learn from the occurrence of events in a different place and time to predict similar trajectories of events. We define Loosely Decoupled Timeseries (LDT) phenomena as two or more events that could happen in different places and across different timelines but share similarities in the nature of the event and the properties of the location. In this work we improve on the use of Recurring Neural Networks (RNN), in particular Long Short-Term Memory (LSTM) networks, to enable AI solutions that generate better timeseries predictions for LDT. We use similarity measures between timeseries based on the trends and introduce embeddings representing those trends. The embeddings represent properties of the event which, coupled with the LSTM structure, can be clustered to identify similar temporally unaligned events. In this paper, we explore methods of seeding a multivariate LSTM from time-invariant data related to the geophysical and demographic phenomena being modeled by the LSTM. We apply these methods on the timeseries data derived from the COVID-19 detected infection and death cases. We use publicly available socio-economic data to seed the LSTM models, creating embeddings, to determine whether such seeding improves case predictions. The embeddings produced by these LSTMs are clustered to identify best-matching candidates for forecasting an evolving timeseries. Applying this method, we show an improvement in 10-day moving average predictions of disease propagation at the US County level.