Invited Speakers



Prof. Yanan Sun
Sichuan University, China

Yanan Sun is currently a professor at the College of Computer Science, Sichuan University, China. He received his Ph.D. degree in computer science from Sichuan University in 2017. From June 2017 to March 2019, he was a postdoctoral fellow at Victoria University of Wellington, New Zealand. His research focuses on evolutionary computation, neural networks, and their applications in neural architecture search. In this research area, he has published 31 peer-reviewed papers including 12 first (correspondence)-authored papers in top IEEE Trans. journals. As PI/Co-PI, he has received five research grants from the Science & Technology Department of Sichuan Province, one from the Chengdu Science and Technology Bureau, and one from the National Natural Science Foundation of China. In 2016, he received the best student paper award of IEEE CIS Chengdu Chapter, National Scholarship of China, and IEEE student travel grant. In 2020, he was awarded as the Innovative talent of science and technology in Sichuan Province (priority funding).

He was invited to be the organizing committee, program committee, special session chair, and tutorial chair of nine international conferences. He was the Thought Leader of Evolutionary Deep Learning from one of the six research focuses established at Victoria University of Wellington. He is the leading organizer of one workshop and two special sessions on the topic of Evolutionary Deep Learning, and the founding chair of IEEE CIS Task Force on Evolutionary Deep Learning and Applications. He is also the Guest Editor of the Special Issue on Evolutionary Computer Vision, Image Processing and Pattern Recognition in Applied Soft Computing, and the Guest Editor of the Special Issue on Evolutionary Deep Neural Architecture Design and Applications in IEEE Computational Intelligence Magazine.

Speech Title: Homogeneous Architecture Augmentation for Neural Predictor
Abstract:
Neural Architecture Search (NAS) can automatically design well-performed architectures of Deep Neural Networks (DNNs) for the tasks at hand. However, one bottleneck of NAS is the prohibitively computational cost largely due to the expensive performance evaluation. The neural predictors can directly estimate the performance without any training of the DNNs to be evaluated, thus have drawn increasing attention from researchers. Despite their popularity, they also suffer a severe limitation: the shortage of annotated DNN architectures for effectively training the neural predictors. In this paper, we proposed Homogeneous Architecture Augmentation for Neural Predictor (HAAP) of DNN architectures to address the issue aforementioned. Specifically, a homogeneous architecture augmentation algorithm is proposed in HAAP to generate sufficient training data taking the use of homogeneous representation. Furthermore, the one-hot encoding strategy is introduced into HAAP to make the representation of DNN architectures more effective. The experiments have been conducted on both NAS-Benchmark-101 and NAS-Bench-201 dataset. The experimental results demonstrate that the proposed HAAP algorithm outperforms the state of the arts compared, yet with much less training data. In addition, the ablation studies on both benchmark datasets have also shown the universality of the homogeneous architecture augmentation.




Assoc. Prof. Wei Shen
Shanghai Jiao Tong University, China

Wei Shen is a tenure-track Associate Professor at the Artificial Intelligence Institute, Shanghai Jiao Tong University, since October 2020. Before that, he was an Assistant Research Professor at the Department of Computer Science, Johns Hopkins University, worked with Bloomberg Distinguished Professor Alan Yuille. He received his B.S. and Ph.D. degrees from Huazhong University of Science and Technology in 2007 and in 2012, respectively. In 2012, he joined Shanghai University, served as an Assistant Professor and then an Associate Professor until Oct 2018. He also worked as a research intern with Prof. Zhuowen Tu at Microsoft Research Asia.

He is an associate editor of Neurocomputing. He also serves as an EAC for VALSE. His research interests lie in the fields of computer vision, machine learning, deep learning, and medical image analysis, particularly in shape based object representation and detection, deep learning algorithms under various learning paradigms and their application to medical image analysis. His research is supported by National Natural Science Foundation of China. He has over 40 peer-reviewed publications in computer vision and machine learning related areas, including IEEE TPAMI, IEEE TIP, IEEE TMI, NIPS, ICML, ICCV, CVPR, ECCV and MICCAI.

Speech Title: Deep Random Forests: Algorithms and Applications
Abstract: Random forests (RFs), or randomized decision trees, are a popular ensemble predictive model, which have a rich and successful history in machine learning in general and computer vision in particular. Deep networks, especially Convolutional Neural Networks (CNNs), have become dominant learning models in recent years, due to their end-to-end manner of learning good feature representations combined with good predictive ability. However, combining these two methods, i.e., Random forests and CNNs, is an open research topic that has received less attention in the literature. A main reason is that decision trees, unlike deep networks, are non-differentiable.

In this talk, I will introduce my recent work on integrating RFs with CNNs (Deep Random Forests) to address various machine learning problems, such as label distribution learning and nonlinear regression. I will show their applications to computer vision problems, such as facial age estimation. I will demonstrate how to learn the Deep Random Forests for different learning tasks by a unified optimization framework.




Assoc. Prof. Kuang Kun
Zhejiang University, China

Kun Kuang, Associate Professor in the College of Computer Science and Technology, Zhejiang University. He received his Ph.D. in the Department of Computer Science and Technology at Tsinghua University in 2019. He was a visiting scholar at Stanford University. His main research interests include causal inference and causally regularized machine learning. He has published over 30 papers in major international journals and conferences, including SIGKDD, ICML, ACM MM, AAAI, IJCAI, TKDE, TKDD, Engineering, and ICDM, etc.

Speech Title: Causal Inference in Observational Studies
Abstract: Causal questions exist in many areas, such as health care, economics, political science, digital marketing, etc. Does a new medication lead to a better performance on a certain illness, compared with the old ones? Does a new marketing strategy improve the sales of a certain products? All these questions can be addressed by the causal inference technique.
The gold standard approaches for causal inference are randomized experiments, for example, A/B testing. However, the fully randomized experiments are usually extremely expensive and sometimes even infeasible. Hence, it is highly demanding to develop automatic statistical approaches to infer causal effect in observational studies.
In this talk, we show some new challenges of causal inference in the wild big data scenarios, including (1) high dimensional and noisy variables, (2) unknown model structure of interactions among variables, and (3) continuous/complex treatment variable. To address these challenges, we proposed Data-Driven Variable Decomposition (D2VD) algorithm, Decomposed Representation Counterfactual Regression (DeR-CFR) model, Differentiated Confounder Balancing (DCB) algorithm, and Generative Adversarial De-confounding (GAD) algorithm. We will show that our proposed algorithms can make a more precise and robust estimation of causal effect in observational studies, compared with the start-of-the-art methods.
 




Assoc. Prof. Yaseen Ahmed Al-Mulla
Sultan Qaboos University, Oman

Dr. Yaseen AI-Mulla is an Associate Professor, Founder and Chair, IEEE GRSS Oman, Scientific Ambassador, IEEE GRSS, Chair, Artificial Intelligence in Remote Sensing & GIS SQU RG, and Director of Remote Sensing and GIS Research Center at Sultan Qaboos University. He received his PhD, BioSystems Engineering, Washington State University, USA. His current research interests include using Remote Sensing technologies for environmental assessment/mapping and land use/change. As well as the use of sensors, instrumentations, UAV (Drones) and IoT for real time monitoring/mapping. Dr. Al-Mulla received 26 national and international awards and recognitions among which he is selected as scientist ambassador representing Oman by the professional association of the Institute of Electrical and Electronics Engineers-Geoscience and Remote Sensing Society (IEEE-GRSS) since 2018. He has professional affiliations with IEEE-GRSS, ASABE, OSE, ICESCO, CHE-ISHS, and MENA-NWC. He formed a formal scientific/research collaboration with more 36 research institutions from 20 countries. He served as Chairman/Member in 56 International, Nationwide and Community level, 41 University level, 48 College level and 39 Departmental level committees. He worked as an Editor-in-Chief for “Al-Hasaad” Outreach Magazine and as a SQU/Oman representative in the Organizing Committee for the second US National Academy and The World Academy of Sciences meetings. Dr. Al-Mulla served as a referee/reviewer for 158 papers for 33 journals /institutions/Book series. His publications include 171 papers, books/book chapters, abstracts, and technical reports/thesis/newspaper articles. He has participated in 74 Conferences, Symposia, and Workshops.

Speech Title: IoT and RS Techniques for Enhancing Water Use Efficiency and Achieving Water Security
Abstract: Water security is among the key elements and necessary to maintain sustainability. The Sultanate of Oman, similar to other arid countries, has natural water resources with limited water recharge capacity to support its present population on natural water and to reduce its water resources deficit. The water security can be attained by the implementation of the state of art modern means of water conservation practice especially in relation with irrigation which consumes 80% -90% of natural water resources. In this paper, the idea of the implementation of automation methods in achieving water security is covered through the control of the water supply to plants in optimal ways and reducing water wastage in farming by using automated closed loop controlled systems, Internet of Things (IOT) and Remote Sensing technology. This approach will not only conserve water but it will also facilitate and give the farmers freedom to control and monitor their farms irrigation remotely just by using smartphones, laptops, tablets and PCs.




Prof. Helder Gomes Cost
Fluminense Federal University, Brazil

Helder Gomes Costa received his PhD in Mechanical Engineering from Pontifícia Universidade Católica do Rio de Janeiro (1994), Brazil. He is currently full professor and the header of the Decision Analysis Group at Universidade Federal Fluminense (UFF), Brazil, acting on the following subjects: decision multicriteria decision making. clustering and performance evaluation. Nowadays, he is the President of the National Association of Post-Graduation and Research in Production Engineering (ANPEPRO), Brazil.

Speech Title: Additive X Non-additive Decision Models: Avoiding Misunderstandings with the Support of Visual OutDeK
Abstract: Decision may be such a complex problem when we deal with different criteria. The complexity of modelling a decision problem rises when the decision maker (DM) takes into account multiple criteria, mainly it them are in conflict or even if there are no standardized ways to measure the alternatives performance under a criterion viewpoint - which is the land where subjectivities grow. There are several Multicriteria Decision Aid Methods (MCDA) methods designed to be applied in a decision situation, each one having its singular features. Because the diversity of decision situations, some methods do not work well in some categories of problems. Despite it, it is usual to apply a decision method even in situation for which such method is not suitable – which provides incoherent decisions. In this speaking, this issue is approached by highlighting the differences between additive and non-additive decision problems, and by describing a non-additive multicriteria decision method. The speaking will be supported by a web app: the Visual OutDeK. The link that follows is suggested to those that wants to know a bit more about the subject approached in the conference: https://doi.org/10.1108/JM2-08-2013-0037.




 

Assoc. Prof. Rasha Ismail
University of Hertfordshire – GAF, Egypt

Rasha Ismail is currently a Programme Leader of Business Administration at University of Hertfordshire-GAF (UH-GAF), Egypt. She also serves as an Associate Professor and moderation coordinator for Business Administration faculty. She received her Ph. D degree in MIS/E-business from University of the West of England, in 2010. From August 2006 to September 2012, she worked in the career of teaching until she got promoted to Assistant Professor at the Arab Academy for Science and Technology and Maritime Transport (AASTMT), Egypt. She joined American university of the Middle East, Kuwait as an Assistant Professor in 2012 and then she got promoted to Associate Professor in 2015 until she left in 2019.

She is a member of International Association of Computer Science and Information Technology (IACSIT) and she was also a Committee Member in International Conference on Computer Technology and Development (ICCTD 2011). She reviewed a number of researches for several conferences.

Her research interests include Process automation and improvements. She has published a number of researches about business process modeling, the transition to e-business and process improvements in education and manufacturing. Rasha Ismail was awarded certificates for attending conferences and presenting papers, she was also awarded a certificate of appreciation as a reviewer from IBIMA, in addition to certificates of appreciation from AUM for her efforts at work.

Speech Title: Disruptive Technology Impacts on Industry
Abstract:
The presentation given by Dr.Ismail talks about the history of Industry Four, as in the three previous revolutions which built up from the mechanized revolution up to the electronic and automation revolution. Furthermore, the definition of Industry Four is given to be a German initiative that encourages the use of high technology production systems that use cyber-physical systems, internet of things and cloud computing for efficiency and competitiveness. The effects of Industry Four are also discussed; these include the improvement in efficiency and quality of production which will lead to an increased competitiveness between manufactures thus achieving the goal of the initiative. Other effects include better, more accurate communication systems, a complete overhaul of manufacturing systems since no producer wants to be left behind, newer business models to adjust to the new production systems and a complete sync between cyber-physical systems and the internet of things.