Inference, Prediction, & Entropy-Rate Estimation of Continuous-Time, Discrete-Event Processes

Inferring models, predicting the future, and estimating the entropy rate of discrete-time, discrete-event processes is well-worn ground. However, a much broader class of discrete-event processes operates in continuous-time. Here, we provide new methods for inferring, predicting, and estimating them. The methods rely on an extension of Bayesian structural inference that takes advantage of neural network's universal approximation power. Based on experiments with complex synthetic data, the methods are competitive with the state-of-the-art for prediction and entropy-rate estimation.

Automated summary

This paper proposes new methods for inferring, predicting, and estimating continuous-time discrete-event processes. The methods are based on an extension of Bayesian structural inference and utilize the universal approximation power of neural networks. Experimental results on complex synthetic data demonstrate that these methods are competitive with the state-of-the-art for prediction and entropy-rate estimation.