Variational inference for Markov jump processes
Manfred Opper and Guido Sanguinetti
Advances in Neural Information Processing Systems 2007
Discrete stochastic processes play an important role in a
large number of application domains. However, realistic systems are
analytically intractable and they have traditionally been analysed using
simulation based techniques, which do not provide a framework for statistical
inference. We propose a mean field approximation to perform posterior
inference and parameter estimation. The approximation allows a practical
solution to the inference problem, while still retaining some important
features of the original problem such as the existence of emerging properties.
We illustrated our approach on two biologically motivated systems.