Stein Variational Online Changepoint Detection with Applications to Hawkes Processes and Neural Networks
This provides a computationally tractable Bayesian approach for detecting changepoints in non-exponential distributions, addressing practical challenges in fields like event modeling and neural networks, though it is incremental as it builds on existing methods.
The authors tackled the problem of online changepoint detection in complex systems by introducing a Stein variational method that generalizes Bayesian online changepoint detection beyond exponential family distributions, applying it successfully to Hawkes processes and LSTM neural networks on real data.
Bayesian online changepoint detection (BOCPD) (Adams & MacKay, 2007) offers a rigorous and viable way to identify changepoints in complex systems. In this work, we introduce a Stein variational online changepoint detection (SVOCD) method to provide a computationally tractable generalization of BOCPD beyond the exponential family of probability distributions. We integrate the recently developed Stein variational Newton (SVN) method (Detommaso et al., 2018) and BOCPD to offer a full online Bayesian treatment for a large number of situations with significant importance in practice. We apply the resulting method to two challenging and novel applications: Hawkes processes and long short-term memory (LSTM) neural networks. In both cases, we successfully demonstrate the efficacy of our method on real data.