The Age of Correlated Features in Supervised Learning based Forecasting

TL;DR

This paper investigates how the freshness of correlated features affects supervised learning forecasting, revealing that training and testing losses are influenced by feature age and proposing methods to improve model performance.

Contribution

It introduces an information-theoretic analysis of feature age impact on forecasting loss and suggests incorporating feature age into training and input features for better accuracy.

Findings

Training loss depends on feature age, not always monotonically.

When data distribution resembles a Markov chain, loss growth with age is approximately non-decreasing.

Joint training on data with varying feature ages improves forecasting performance.

Abstract

In this paper, we analyze the impact of information freshness on supervised learning based forecasting. In these applications, a neural network is trained to predict a time-varying target (e.g., solar power), based on multiple correlated features (e.g., temperature, humidity, and cloud coverage). The features are collected from different data sources and are subject to heterogeneous and time-varying ages. By using an information-theoretic approach, we prove that the minimum training loss is a function of the ages of the features, where the function is not always monotonic. However, if the empirical distribution of the training data is close to the distribution of a Markov chain, then the training loss is approximately a non-decreasing age function. Both the training loss and testing loss depict similar growth patterns as the age increases. An experiment on solar power prediction is conducted to validate our theory. Our theoretical and experimental results suggest that it is beneficial to (i) combine the training data with different age values into a large training dataset and jointly train the forecasting decisions for these age values, and (ii) feed the age value as a part of the input feature to the neural network.