International Conference on Advancements in Computational Sciences

19-21 February 2018, Lahore Pakistan




Profile: Dr. Amal El Fallah Seghrouchni

Title: Towards the Development of Intelligent and Interactive Systems The ambient Intelligence context

Abstract:

The effervescence around the Artificial Intelligence today raises many challenges. Beyond the ethical and regulatory issues, the technical and scientific challenges remain major in particular because of the end-user’s expectations. We observe the emergence of new topics to deal with the new environments of deployment and the various modes of interaction with end-users. Indeed, the computational systems in the context of ambient intelligence become increasingly open to users which require robustness and autonomy. More and more, technical systems penetrate our daily lives and start to meet humans as daily assistants or companions, setting a fundamental need to incorporate ambient intelligence advancements.

This talk will address the question how to develop intelligent and interactive systems in the context of ambient intelligence. It will introduce the Ambient Intelligence Context, the requirements of users in terms of adaptation and context awareness. Then it will present the multiagent system paradigm as an interesting way to meet with user’s requirements. Finally, the talk will present some of our contributions to this domain of research.




Profile: Dr. Yasar Ayaz

Title: Robotics on the R.I.S.E.

Abstract:

Robotics on the R.I.S.E.

Robotics is one of the most revolutionizing emerging technologies of the modern era and is widely expected to spearhead the upcoming technological revolution where robots and intelligent machines will become an integral part of everyday human life. Ranging from area specific domains such as industrial automation and surgical robotics to machines that operate in common everyday human households, robots can already be seen assisting humans in virtually every aspect of their lives. Not only do they help reduce human effort but also, at the same time, improve the precision and accuracy with which these tasks are accomplished. Recent years have seen almost every major corporation invest into Robotics. In Pakistan also, several research organizations and companies are working in areas relating to Robotics. Among these, ths NUST-SMME Robotics and Intelligent Systems Engineering (R.I.S.E.) lab is one of the most accomplished. This talk will focus on the developments of Robotics in Pakistan and abroad, with a key focus on Robotics research activities being conducted in Pakistan by R.I.S.E. lab.




Profile: Dr. Ashiq Anjum

Topic: Blockchain enabled Distributed Systems

Abstract:

A blockchain is a decentralized and distributed digital ledger that is used to record transactions across distributed systems. Each ledger entry is time-stamped and permanent so that the records cannot be altered retroactively without the alteration of all subsequent blocks. These ledgers (blockchains) are characterized by four key elements: shared distribution (every computer node has the same version which is updated simultaneously), privacy (access can be based on granted permission), consensus (all parties on the blockchain agree to the validity of any given transaction), and smart contracts (a computer protocol intended to enforce the transaction according to present conditions).

There are multiple blockchain technology providers. The available technologies exhibit a tradeoff between performance and scalability, incorporate public and private models, and support permissioned and permission-less ledgers.

Systems and applications of blockchain in healthcare have already been envisioned including exchange of healthcare data, including effort and trust when accessing data from multiple sources, the high cost and latency of healthcare data exchange, and the burden of traceability in clinical validation studies. Patients and healthcare providers would have immediate access to accurate and up-to-date patient records. As such, blockchain may represent an emerging force in healthcare data management.

This talk will provide a critical review of block chain technology, infrastructure and applications from the perspective of Healthcare.




Profile: Dr. Mohammad Mousavi

Title: Model-Based Testing Cyber-Physical Systems: Theory and Practice

Abstract:

Cyber-physical systems are the results of the tight integration of computation, control, and communication. Hence, their mode-based design involves hybrid and multi-disciplinary modelsaddressing discrete behaviour, integrated with continuous dynamics along with (often asynchronous) communication. In this talk, we focus on model-based conformance testing of such systems and present an overview of the theories of conformance testing adapted to this domain. Then, we present some of our ongoing research results towards a practical framework for model-basedtesting cyber-physical systems. Based on joint work with: A. Aerts (TU Eindhoven, NL), H. Araujo (F.U. Pernambuco, Brazil), G. Carvalho (F.U. Pernambuco, Brazil), M. Gazda (Leicester, UK), N. Khakpour (Linneaus U., Sweden), M. Mohaqeqi (Uppsala, Sweden), M.A. Reniers (TU Eindhoven, NL), A. Sampaio (F.U. Pernambuco, Brazil), M. Taromi Rad (Halmstad, Sweden)




Profile: Dr. Jan Friso Groote

Title: Model checking and model based software engineering as an effective boost for software quality.

Abstract:

In this talk we explain how model checking and model based software engineering are effective tools if it comes to the modern construction of software. Model based software engineering use restricted models for particular domains in which software entities are described. We are mainly interested in languages from the realm of automata theory that are particularly suited for the control aspect of embedded systems. From the models code is generated automatically. Although the models are a huge advantage because they are much more concise, it is also difficult to write them in a completely correct way. To increase the level of correctness we write requirements, often in the form of modal mu-calculus formulas, using which we express essential aspects of the behaviour that the models must have. These are checked formally. In field experiments, carried out at Philips Healthcare, we found that the number of bugs reported during development went down with up to a factor 10 and the development time was always reduced, sometimes even with a factor 3. As it stands multiple industries now slowly take model checking and model based software engineering up as their standard way of working.




Profile: Dr. Mark Hoogendoorn

Title: Machine Learning for Health

Abstract:

In the domain of health, a wealth of data is available about the health state and health behavior of people. This data is not only limited to electronic medical records, but relevant data also includes data collected through smart devices such as mobile phones and smart watches. The amount of data is overwhelming, and traditional techniques in the medical domain are no longer adequate to fully exploit this data. Machine learning techniques can help out, for instance for predictive modeling (e.g. estimate the risk to develop a certain disease) or personalization of treatments (provide the treatment best suited for a particular patient). Applying machine learning to health data is however far from trivial. In this presentation, I will discuss the research that my colleagues and I have done in this area. Topics I will discuss involve the automated identification of features from raw health data, the development of predictive modeling techniques for the health domain, and techniques that personalize treatments towards users. Examples will be taken from the intensive care unit, cancer and the domain of mental health.

Financial Sponsors Technical Sponsors
International Conference on Advancements in Computational Sciences
-