2023 3rd International Conference on Artificial Intelligence, Automation and High Performance Computing (AIAHPC 2023)
Count Down

Speakers

KEYNOTE SPEAKER I IN AIAHPC 2022

pa116.png

A. Prof. Pavel Loskot

Zhejiang University-University of Illinois at Urbana-Champaign Institute (ZJUI), China

Research Interests: models, methods and algorithms for probabilistic and statistical inference, Monte-Carlo simulations and related signal and data processing problems

Biography:

Pavel Loskot joined the ZJU-UIUC Institute in January 2021 as the Associate Professor after being nearly 14 years with Swansea University in the UK. He received his PhD degree in Wireless Communications from the University of Alberta in Canada, and the MSc and BSc degrees in Radioelectronics and Biomedical Electronics, respectively, from the Czech Technical University of Prague in the Czech Republic. He is the Senior Member of the IEEE, Fellow of the Higher Education Academy in the UK, and the Recognized Research Supervisor of the UK Council for Graduate Education. His current research interest focuses on problems involving statistical signal processing and importing methods from Telecommunication Engineering and Computer Science to other disciplines in order to improve the efficiency and the information power of system modeling and analysis. Dr. Pavel is a senior member of the IEEE.


Title & abstract:

Towards Interpretable Computing: Processing Event Time Series to Understand Models of Dynamic Systems


Traditional analysis of models of dynamic systems relies on processing synthetic data generated by stochastic simulations of these models. This is a general approach which is often also very suitable for processing the measurements from real-world systems. However, stochastic simulations offer, for free, a lot more information to consider, and to gain much better insights about the simulated models. Here, it is proposed to record changes in the global state of the modeled system in the course of its simulations as a time series of defined events represented by numerical or categorical variables. The overlapping or non-overlapping event sub-sequences can be transformed into event sets or multi-sets. It enables sets-calculus, and defining unions and intersections based set-distances in the matrix profile analysis. Furthermore, the event sets can be used to find causal relationships which are newly defined here in terms of the nearly certain and nearly uncertain conditional event sets. The proposed analytical framework of event time series is demonstrated for stochastic simulations of biochemical reaction networks.



KEYNOTE SPEAKER III IN AIAHPC 2022

mohand.png

Dr. Mohamed Abdellatif

School of civil engineering, University of Leeds, UK

Research Interests: Computer vision, Machine learning

Biography:

Mohamed Abdellatif received PhD in Computer Vision from Okayama University in Japan. His research interests include computer vision, Machine learning and visual SLAM. He worked in the academia  and Industry in several countries. He had experience in curriculum design and reform in several universities. He is now a research fellow in The University of Leeds working in vision solutions for road defect monitoring and assessment. He serves as technical reviewer for several computer vision journals and conferences. Dr. Abdellatif is a member of the IEEE.


Title & abstract:

Deep learning and Computer vision research for autonomous inspection of Infrastructure and smart cities monitoring


There is a trend to exploit computer vision and deep learning for inspection of infrastructure in smart cities such as bridges and other civil structures. The challenges of using drones or mobile robot to inspect and detect anomalies are huge but is worth exploring. The recent challenges that deep learning face is the need for enormous and annotated data to learn the visual information. Deep learning replaced mobel - based geometry approaches but still much can be learned from the geometry models. The geometry can give us good hints about metrics needed to select fewer training data for deep learning. Carefully selected training data spanning wider range of the required metric will enable the deep learning model to generalize faster after training with a small number of images.


 



KEYNOTE SPEAKER III IN AIAHPC 2022

Julien.jpg

Dr. Julien LE KERNEC

James Watt School of Engineering, University of Glasgow, UK

School of Information and Communication, University of Electronic, Science and Technology of China, Chengdu, China

Information and Signal Processing Team, University Cergy-Pontoise, France

Research Interests: Radar system design, software defined radio/radar, signal processing, and health applications


Title & abstract:

Radar sensing in animal welfare


In this seminar, I will discuss the place of radar sensing in precision farming. First, the importance of lameness detection in dairy cows and the urgency to address welfare in farming. The second part will give an overview of existing sensing modalities for lameness detection and explains why radar is to address this issue. The third section presents developments in machine learning that helps improve performances in classification. Finally, I will conclude with open challenges.





Inviting more...