ProPPA is a probabilistic programming language for continuous-time dynamical systems, developed as an extension of the stochastic process algebra Bio-PEPA. It offers a high-level syntax for describing systems of interacting components with stochastic behaviours where some of the parameters are unknown. Such systems occur in many and diverse fields, including biology, ecology and urban transport, and while their modelling and analysis are important, existing methodologies are often not accessible to non-experts, in addition to being tailor-made rather than generally applicable. In particular, parameter learning can be of crucial significance but is a difficult problem due to the complexity of the underlying probabilistic model — namely, the continuous time setting and the fast-growing number of states.
The purpose of the ProPPA framework is to facilitate both the description of these systems and the process of inference, by automating the application of appropriate algorithms. The language is equipped with different parameter inference methods, including a novel MCMC scheme which employs a random truncation strategy to obtain unbiased likelihood estimates. This method is particularly suited to systems with infinite state-spaces, which were previously not manageable without imposing an ad-hoc truncation. Other methods include a naive Approximate Bayesian Computation algorithm and a sampler based on a continuous approximation of the state-space.
The workshop was held on January 11th and 12th.
Logic has proved in the last decades a powerful tool in understanding complex systems. It is instrumental in the development of formal methods, which are mathematically based techniques obsessing on hard guarantees. Learning is a pervasive paradigm which has seen tremendous success recently. The use of statistical approaches yields practical solutions to problems which yesterday seemed out of reach. These two mindsets should not be kept apart, and many efforts have been made recently to combine the formal reasoning offered by logic and the power of learning.
The goal of this workshop is to bring together expertise from various areas to try and understand the opportunities offered by combining logic and learning.
There are 12 invited speakers and a light programme (less than 5h per day) so as to give enough time for discussions.
Add comment