User-Dependent Neural Sequence Models for Continuous-Time Event Data

Continuous-time event data are common in applications such as individual behavior data, financial transactions, and medical health records. Modeling such data can be very challenging, in particular for applications with many different types of events, since it requires a model to predict the event types as well as the time of occurrence. Recurrent neural networks that parameterize time-varying intensity functions are the current state-of-the-art for predictive modeling with such data. These models typically assume that all event sequences come from the same data distribution. However, in many applications event sequences are generated by different sources, or users, and their characteristics can be very different. In this paper, we extend the broad class of neural marked point process models to mixtures of latent embeddings, where each mixture component models the characteristic traits of a given user. Our approach relies on augmenting these models with a latent variable that encodes user characteristics, represented by a mixture model over user behavior that is trained via amortized variational inference. We evaluate our methods on four large real-world datasets and demonstrate systematic improvements from our approach over existing work for a variety of predictive metrics such as log-likelihood, next event ranking, and source-of-sequence identification. Recent work in machine learning has sought to address these limitations via the use of deep recurrent neural networks (RNNs). These models, such as Du et al. [2016] , use expressive representations for the intensity function, use event embeddings to avoid parameter explosion, and optimize the associated log-likelihood via stochastic gradient methods. A variety of approaches have been explored to address the complex mix of discrete events and continuous time that occur in real-world event 34th Conference on Neural Information Processing Systems (NeurIPS 2020),