Approximate Probabilistic Inference for Time-Series Data A Robust Latent Gaussian Model With Temporal Awareness

TL;DR

This paper introduces tDLGM, a robust probabilistic generative model for non-stationary time series data that effectively captures temporal dependencies and withstands data errors, outperforming traditional models.

Contribution

The paper proposes a novel Time Deep Latent Gaussian Model (tDLGM) with a unique architecture and regularizer for robust time series modeling and inference.

Findings

tDLGM accurately reconstructs complex time series.

tDLGM demonstrates robustness against noise and faulty data.

The model outperforms traditional methods in capturing temporal relationships.

Abstract

The development of robust generative models for highly varied non-stationary time series data is a complex yet important problem. Traditional models for time series data prediction, such as Long Short-Term Memory (LSTM), are inefficient and generalize poorly as they cannot capture complex temporal relationships. In this paper, we present a probabilistic generative model that can be trained to capture temporal information, and that is robust to data errors. We call it Time Deep Latent Gaussian Model (tDLGM). Its novel architecture is inspired by Deep Latent Gaussian Model (DLGM). Our model is trained to minimize a loss function based on the negative log loss. One contributing factor to Time Deep Latent Gaussian Model (tDLGM) robustness is our regularizer, which accounts for data trends. Experiments conducted show that tDLGM is able to reconstruct and generate complex time series data, and that it is robust against to noise and faulty data.