Personal tools
You are here: Home Events Lab Lunch Talk by Anastasis Georgoulas

Lab Lunch Talk by Anastasis Georgoulas

— filed under:

Machine Learning for Formal Dynamical Systems

What
  • Lab Lunch
When Nov 11, 2014
from 01:00 PM to 02:00 PM
Where MF2
Add event to calendar vCal
iCal

 

Formal methods, such as process algebras, have been demonstrated to be useful frameworks for the modelling and analysis of complex systems. Finding parametrizations for these models that match expected or observed behaviour is an open problem, however, with solutions often being time-consuming and ad-hoc. Algorithms from the field of machine learning, on the other hand, can be used to extract information from observed data and manage uncertainty in models in a statistically consistent way. I will be talking about our efforts to bring together the two fields, including ProPPA, a probabilistic programming process algebra that incorporates uncertainty in the description of the system and its semantics. This is joint work with Jane Hillston and Guido Sanguinetti.

Document Actions