Temporal Multimodal Multivariate Learning

TL;DR

This paper presents a novel temporal multimodal multivariate learning framework that improves decision-making in complex, uncertain, and time-dependent scenarios by leveraging correlations across multiple variables and time stages, demonstrated on real-world datasets.

Contribution

It introduces a new decision-making model that learns and transfers information across multiple outcomes and time, addressing complex uncertainties in dynamic environments.

Findings

Outperforms baseline prediction methods on urban traffic data

Achieves superior accuracy in hurricane ensemble forecasting

Demonstrates effectiveness in real-world, multi-variable, time-dependent tasks

Abstract

We introduce temporal multimodal multivariate learning, a new family of decision making models that can indirectly learn and transfer online information from simultaneous observations of a probability distribution with more than one peak or more than one outcome variable from one time stage to another. We approximate the posterior by sequentially removing additional uncertainties across different variables and time, based on data-physics driven correlation, to address a broader class of challenging time-dependent decision-making problems under uncertainty. Extensive experiments on real-world datasets ( i.e., urban traffic data and hurricane ensemble forecasting data) demonstrate the superior performance of the proposed targeted decision-making over the state-of-the-art baseline prediction methods across various settings.