Invited Speaker

Invited Speech I ( online )

Prof. Boudour Ammar,
National Engineering School of Sfax (ENIS),Tunisia

Advanced Deep Learning with & Swarm optimization 


Deep Learning is a subset of Artificial Intelligence, which directs computing to perform classification tasks directly from texts, images, or signals. Deep Learning is one of the most popular domains in the AI space, allowing to develop multi-layered models of varying complexities. The term deep refers to the number of hidden layers in the network. For optimal results, Deep Learning requires large amounts of data and substantial computing power.

In this talk, we will see the foundations of Deep Learning, we understand how to build an advanced deep neural network and learn how to lead successful machine learning.  Most methods of Deep Learning are on neural network architectures especially the optimization techniques.

The swarm intelligence as an optimization technique can achieve a complex and intelligent behavior through the local interactions between its members. The algorithms of swarm intelligence are integrated to deep learning to optimize the corresponded parameters.  

Optimized Deep Learning has its applications in the fields of Automated Driving, Image Recognition, Emotion and Fraud Detection, Natural Language Processing, Healthcare, Security, Personalized Services, and more.



Boudour Ammar is currently an assistant professor with the Department of Computer Engineering and Applied Mathematics at National Engineering School of Sfax (ENIS). She was born in Sfax, Tunisia. She graduated in computer science in 2005. She received the Master degree in automatic and industrial computing from the National School of Engineers of Sfax, University of Sfax in 2006. She obtained a PhD degree in recurrent neural network learning model for a biped walking simulator with the Research Group on Intelligent Machines (REGIM), University of Sfax, since February 2014.
Her research interests include iBrain (Artificial neural networks, Machine learning, Recurrent neural Network) & i-health (Autonomous Robots, Intelligent Control, medical applications, EEG and ECG signals).

In recent years, Dr. Boudour published many highly cited research papers in IEEE Transaction of neural Networks and Learning systems, IEEE Transactions on Affective Computing, Neural Processing Letters, Applied Soft Computing, Neurocomputing, Cybernetics and systems journals.  She also published papers in conferences such as International Joint Conference on Neural Networks (IJCNN), International Conference on Neural Information Processing, IEEE International Conference on Systems, Man, and Cybernetics.
She, currently, serves as a reviewer of some international conferences and journals like Neural Computing and IEEE TNNLS.

She also served in several volunteering positions as a head of the Career Center and Certification Skills 4C-ENIS, IEEE WIE Tunisia Affinity Group chair, chair and Vice-chair of IEEE Computational Intelligence Societies & IEEE RAS Tunisia and YP active Member. She organized the Robocomp robotics competition and the World Robot Olympiad Tunisia Qualification. She served as an organizing member in several workshops and international conferences including Workshop on Intelligent machines Theory & Applications (WIMTA), Workshop in Intelligent Decision Support Systems (iDSS), Multi-Conference on Systems, Signals and Devices (SSD).