Instructors/Speakers Prof. Ponnuthurai Nagaratnam SUGANTHAN Nanyang Technology Univeresity Singapore Abstract This talk will first introduce the main non-iterative learning paradigms such as the randomization based feedforward neural networks (e.g. random vector functional link from 1994, extreme learning machine from 2004), random forest, and kernel ridge regression. Some of these non-iterative methods have closed form solutions enabling them to be trained extremely fast. The talk will highlight the similarities and differences among these methods developed over the last 25 years. The talk will also present benchmarking studies of these methods using classification and forecasting datasets. Biography Professor Ponnuthurai Nagaratnam Suganthan (or P N Suganthan) received the B.A degree, Postgraduate Certificate and M.A degree in Electrical and Information Engineering from the University ...

Instructors/Speakers Prof. Anthony T.S. HO University of Surrey Abstract This talk will present an overview of the theory and applications of Benford's law for anomaly detection in natural data. Some examples will be highlighted including the detection of glare effect in images and classification of biometric images for privacy protection, as well as security attacks related to network traffic data. Recent research based on this law has further shown that consistent anomaly patterns could be achieved for different network attacks, leading to the potential identification/pattern recognition of various types of attacks. Moreover, Benford's law has also been successfully applied for the detection of Alzheimer’s Disease based on Electroencephalogram (EEG) data and this will be highlighted in the presentation. Biography Professor ...

Instructors/Speakers Prof. David Kaber North Carolina State University (NCSU) Abstract Recent research in intelligent systems has discussed the characteristics of autonomous systems. This same work has evaluated automated systems in terms of the understanding of autonomy. This situation has led to confusion of automation technology and autonomous agents. In this talk, I will differentiate the concepts of automation and autonomy with a new framework of agents. The framework is complemented by observations on characteristics of automated vs. autonomous systems, identification of error and failure modes, and formulation of a matrix of design constraints dictating possible applications of each type of agent. I will also discuss levels of system automation along with types of autonomy. A definition of autonomy will be ...

https://www.fst.um.edu.mo/wp-content/uploads/2018/06/sem20180621_01.jpg

Instructors/Speakers Prof. Zixiang XIONG Department of Electrical and Computer Engineering Texas A&M University Abstract In the past decade, much progress has been made in image denoising due to the use of low-rank representation and sparse coding. In the meanwhile, state-of-the-art algorithms also rely on an iteration step to boost the denoising performance. However, the boosting step is fixed or non-adaptive. In this work, we perform rank-1 based fixed-point analysis, then, guided by our analysis, we develop the first adaptive boosting (AB) algorithm, whose convergence is guaranteed. Preliminary results on the same image dataset show that AB uniformly outperforms existing denoising algorithms on every image and at each noise level, with more gains at higher noise levels. Biography Zixiang Xiong received ...