Temporal Variational Implicit Neural Representations

2025-06-02

AI Summary

Irregular multivariate time series pose significant challenges for personalized missing-value imputation and forecasting under low-data regimes. Method: This paper proposes a continuous-time generative framework that integrates implicit neural representations with variational latent variable modeling. The approach conditions the continuous-time generative function distribution on signal-specific covariates, enabling single-pass forward inference for personalized imputation and forecasting—without fine-tuning, meta-learning, or repeated training. A covariate-conditioned latent space design allows a single unified model to support diverse imputation and forecasting tasks. Results: Evaluated on real-world sparse time-series datasets, the method reduces imputation error by an order of magnitude compared to state-of-the-art baselines, achieving substantial gains in both accuracy and computational efficiency. It is particularly effective in few-shot and highly sparse settings, demonstrating robust generalization from limited observations.

Technology Category

Machine Learning: Time-Series/Data StreamsNatural Language Processing: GenerationCognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalizationSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingGraph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

We introduce Temporal Variational Implicit Neural Representations (TV-INRs), a probabilistic framework for modeling irregular multivariate time series that enables efficient individualized imputation and forecasting. By integrating implicit neural representations with latent variable models, TV-INRs learn distributions over time-continuous generator functions conditioned on signal-specific covariates. Unlike existing approaches that require extensive training, fine-tuning or meta-learning, our method achieves accurate individualized predictions through a single forward pass. Our experiments demonstrate that with a single TV-INRs instance, we can accurately solve diverse imputation and forecasting tasks, offering a computationally efficient and scalable solution for real-world applications. TV-INRs excel especially in low-data regimes, where it outperforms existing methods by an order of magnitude in mean squared error for imputation task.

Problem

• Modeling irregular multivariate time series efficiently
• Enabling individualized imputation and forecasting accurately
• Outperforming existing methods in low-data regimes

Innovation

• Probabilistic framework for irregular time series
• Integrates neural representations with latent variables
• Efficient single-pass individualized predictions