Dr Yasmin Rafiq

Email: dr.rafiqy@gmail.com

Birkbeck University of London (Present)

I am currently holding a Research Associate position at Birkbeck University of London. The title of my project is: Perturbation Analysis for Probabilistic Verification. The main focus of this project is on probabilistic model checking, which is used to verify stochastic systems against various quantitative properties, for instance, reliability, security and performance. Current practice of probabilisitic model checking usually assumes that numerical quantities (e.g., transition probabilities, transition rates) in stochastic models are known exactly, or can be acquired precisely. However, real-world systems, for instances those from engineering, biology, or economics, are governed by parameters whose values must be empirically estimated. These values are of statistical nature, therefore are subject to perturbations, which raises the sensitivity or robustness issues of the verification results. To address this issue, the project will carry out perturbation analysis, i.e., to analyse how the verification result is affected by the perturbed parameters of the model and provide a quantitative measure, in terms of various perturbation bounds. The project will also develop efficient and effective algorithms to compute these perturbation bounds, and identify their computational complexity.

Past Positions:

  • Research Associate at Imperial College London (May 2015 – April 2017 )

Project title: Application of Machine Learning to Adaptive Systems for Privacy Dynamics. I worked on a EPSRC research project on "Privacy Dynamics: Learning from the Wisdom of Groups" where I enjoyed collaborating with researchers at Open University,Exeter University and at Imperial College London. I was the key contributor in developing a dynamic privacy control framework, that helps social network users to make informed decisions about their social network interactions in a way that facilitates their social benefit gain whilst reducing their risk of privacy breach. Our approach enables the incorporation of more nuanced privacy controls into OSNs. Specifically, we propose a reusable, adaptive software architecture that uses rigorous runtime analysis to help OSN users to make informed decisions about suitable audiences for their posts. This is achieved by supporting dynamic formation of recipient-groups that benefit social interactions while reducing privacy risks. We exemplify the use of our approach in the context of Facebook. This work will be published in the In Proceeding of the Automated Software Engineering (ASE), 2017 under the category 'New Ideas'.

Education:

  • PhD Computer Science at University of York (2010-2014)

My PhD thesis introduced new techniques for learning the parameters and the structure of discrete-time Markov chains, a class of models that is widely used to establish key reliability, performance and other QoS properties of real-world systems. The new learning techniques use as input run-time observations of system events associated with costs/rewards and transitions between the states of a model. When the model structure is known, they continually update its state transition probabilities and costs/rewards in line with the observed variations in the behaviour of the system. In scenarios when the model structure is unknown, a Markov chain is synthesised from sequences of such observations. The two categories of learning techniques underpined the operation of a new toolset for the engineering of self-adaptive service-based systems, which was developed as part of this research. The thesis introduces this software engineering toolset, and demonstrates its efectiveness in a case study that involves the development of a prototype telehealth service-based system capable of continual self-verification. Link to PhD thesis

  • MRes Photonic Network at Aston University (2009-2011)
  • BSc Hons Computer Science at Aston University (2005-2009)

Research Interest:

I have extensive experience in model-driven engineering, quantitative verification and autonomous computing. My particular interests are in self-adaptive, self-healing and self-managing systems that have capabilities to adapt, reconfigure and verify system requirements at runtime during unpredictable changes, in workload, requirements and environment. My particular focus is on distributed systems that are deployed in Web Architecture. My PhD research was focused on online model learning for quality-of-service engineering, with applications to the verification of non-functional requirements for adaptive computer systems, analysis of non-functional requirements of complex software systems, and service-based architectures.

Publication:

  • Y.Rafiq, L.Dickens, A.Russo, A.K. Bandara, M. Yang, A. Stuart, M. Levine, G. Calikli, B.A. Price, and Bashar Nuseibeh. Learning to Share: Engineering Adaptive Decision-Support for Online Social Networks. In Proceeding of the Automated Software Engineering (ASE), 2017.
  • R. Calinescu, C. Ghezzi, K. Johnson, M. Pezzé, Y. Rafiq and G. Tamburrelli, "Formal Verification With Confidence Intervals to Establish Quality of Service Properties of Software Systems," in IEEE Transactions on Reliability, vol. 65, no. 1, pp. 107-125, March 2016 doi: 10.1109/TR.2015.245293
  • R. Calinescu, Y.Rafiq, K. Johnson, and M. E. Bakir. Adaptive model learning for continual verification of non-functional properties. In 5th ACM/SPEC International Conference on Performance Engineering (ICPE), 2014.
  • R. Calinescu, K. Johnson, and Y. Rafiq. Developing self-verifying service-based systems. In Automated software Engineering (ASE), 2013 IEEE/ACM 28th International Conference on, page 734-737, 2013.
  • R. Calinescu and Y. Rafiq. Using intelligent proxies to develop self-adaptive service-based systems. In Theoretical Aspects of Software Engineering (TASE), International Symposium on pages, 131-134, July 2013.
  • R. Calinescu, K. Johnson, Y. Rafiq, S. Gerasimou, G. Costa Silva, and S. N. Pehlivanov. ``Continual Verification of Non-Functional Properties in Cloud-Based Systems." In NiM-ALP@ MoDELS, pp. 1-5. 2013.
  • R.Calinescu, K. Johnson, and Y. Rafiq. Using observation ageing to improve Markovian model learning in QoS engineering. In Proceeding of the second joint WOSP/SIPEW international conference on Performance engineering, ICPE'11, pages 505-510, New York, NY, USA, 2011. ACM.