Xiaobo Zhang
Title: GRA-CNN: A Novel Convolutional Neural Network Method with Gray Relational Analysis
Bio:
Xiaobo Zhang was born in Suizhou, China in 1977. Master of Computer Science, Doctor of Engineering (Control Theory and Control Engineering), Now working in the Internet of Things Department of Automation, Guangdong University of Technology. Guangdong Science and Technology Commissioner. Engaged in teaching, technology research and development, product design and scientific theoretical research in the field of industrial control and information technology services. His main research interests are hardware and software design for IoT and big data, intelligent information processing and intelligent control (robot) and other industrial control systems. He has participated in more than 60 national and provincial research projects and corporate research projects, including National Natural Science Foundations, Natural Science Foundations of Guangdong Province, major science and technology projects in Guangdong Province, and a combination of production, education and research projects of the Ministry of Education of Guangdong Province etc. In recent years, there have been more than 30 papers published in academic journals with important influence at home and abroad, including 3 books as chief editor. He applied for more than 40 invention patents and obtained 6 software copyrights.
Abstract:
Deep convolutional neural network is one of the most important deep learning models, widely used in image recognition, speech recognition and other fields. In the research and application of convolutional neural networks, the number of hidden layers and nodes is generally determined by experience and expert knowledge. In this sense, achieving optimal network architecture with excellent generalization capabilities and low model complexity is time consuming and even impossible. In order to solve this problem, this paper proposes a gray correlation analysis method to optimize the convolutional neural network (GRA-CNN). The method evaluates the correlation between hidden nodes and output nodes in the network model and selects hidden nodes with higher correlation. Deleting hidden nodes with smaller degrees ensures that the final convolutional neural network can have better recognition accuracy while maintaining higher training performance and higher system stability and adaptability. Experiments on several benchmark datasets show that the proposed method is superior to many other methods in terms of model performance or model complexity.
Reza Langari
Title: An Approach to Shared Control for Automated Driving
Bio:
Renza Langari is Professor of Mechanical Engineering and JR Thompson Department Head Chair, Engineering Technology and Industrial Distribution in the College of Engineering at Texas A&M University. Dr. Langari received the B.Sc., M.Sc., and Ph.D. degrees from the University of California, Berkeley, CA, USA, in 1981, 1983, and 1991, respectively. He was with Measurex Corp. (1984-1985); Integrated Systems, Inc. (1985-1986).; and Insight Development Corporation (1987-1989) prior to starting his academic career at Texas A&M University in 1991. He has since held research positions at NASA Ames Research Center, Rockwell International Science Center, United Technologies Research Center, and the U.S. Air Force Research Laboratory. Dr. Langari’s expertise is in the area of computational intelligence, with application to robotics and autonomous systems. He is the author/co-author of four books and over two hundred technical papers.
Abstract:
This seminar presents a solution to decision making for automated driving based on a combination of game theoretic model predictive control (GTPMC) and reinforcement learning (RL). We first present the rationale for shared autonomy (as opposed to fully automated driving) followed by a framework for assignment of control authority to the machine (i.e. vehicle automation) wherein game theory is used to represent the interaction between the human and the machine. This framework is initially evaluated in simple traffic situations where the vehicle is subjected to discretionary and mandatory lane changes. This initial study demonstrates the general feasibility of the proposed approach via a series of simulated case studies. We will subsequently discuss the use of reinforcement learning to capture the interaction of vehicles in more complex traffic situations. The resulting model is then incorporated in the vehicle automation which is then combined with the game theoretic control authority assignment to investigate the proposed shared autonomy framework in relatively complex highway driving settings. Once again, a series of simulated case studies demonstrate the efficacy of the proposed approach while also highlighting the path towards further development of this framework.
Neha Sharma
Title: Data To Data Product
Bio:
Neha Sharma is a data science crusader who advocates its application for achieving sustainable goals, solving societal, governmental and business problems as well as promotes the use of open data. She has more than 24 years of experience and presently working with Tata Consultancy Services and is a Founder Secretary, Society for Data Science. Prior to this she has worked as Director of premier Institute of Pune, that run post-graduation courses like MCA and MBA. She is an alumnus of a premier College of Engineering and Technology, Bhubaneshwar and completed her PhD from prestigious Indian Institute of Technology, Dhanbad. She is a Senior IEEE member, ACM Distinguished Speaker and was Secretary – IEEE Pune Section for the year 2021 and 2022. She is an astute academician and has organized several national and international conferences and published several research papers. She is the recipient of “Best PhD Thesis Award” and “Best Paper Presenter at International Conference Award” at National Level. Recently, she has received the Golden Book Awards 2023 for the book titled “Towards Net-Zero Targets: Usage of Data Science for Long-Term Sustainability Pathways”. She is a well-known figure among the IT circle, and well sought over for her sound knowledge and professional skills. Neha Sharma has been instrumental in integrating teaching with the current needs of the Industry and steering students towards their bright future.
Abstract:
The APIs published by Netflix, LinkedIn, Twitter, Facebook etc are examples of data products. The keynote would emphasize on the journey of data from being raw to getting converted to data products. Data products are typically referred to in the business space, meaning any application of data that is of value to the business. Typical data products are predictive, descriptive or prescriptive models, as well as insights. Those data products that can help a business generate revenue, optimize costs, mitigate risk, improve compliance… make up a large part of what the whole data value chain – from data sourcing to value creation – should strive to achieve. After all, data, its associated effort and its associated technology investment is all a waste of time and money if it does nothing for the organization’s stakeholders. The lecture would focus on how raw data, after pre-processing, applying statistical analysis and building models, gets converted into a fully developed product.
Dr. Deepak Garg
Title: Opportunities and Challenges in Generative AI
Bio:
Deepak Garg is currently working as Vice Chancellor SR University and is Director Leadingindia.ai Delhi, India. Prior to that he was Dean, School of Computer Science Engineering; Dean International Relations and Corporate Outreach, Bennett University, Greater Noida. With 24 years of rich experience, he is leading the largest Development, Skilling and Research initiative in AI in India with more than 1000 institutional collaborators. He is a chief consultant for algorithmguru.in. He did his Ph.D. in efficient algorithm design in 2006. He possess an innate flair of executing research projects in the field of Data Science, Data Analytics, Deep Learning, Advance Data Structures and Algorithms, MOOCs, Transformations in higher education with quality perspective. He served as chair of IEEE Computer Society, India IEEE Education Society (2013-15).
He is a Logical thinker and strong communicator with noted success of 20 SCI (Web of Science) publications and 700 million INR funding in Grants. He has 1400+ citations count and h-index of 18 in Google Scholar. Credit of having 40 Scopus papers and in total 110+ publications including IEEE Transactions, Springer and Elsevier Journals. Worked in national and international committees on NBA, NAAC accreditation and is only ABET PEV from India till now. Known as algorithm gurus in India and created a free resource as algorithmguru.com. He is an active Blogger with Times of India with nickname as “Breaking Shackles”. His website www.gdeepak.com gives more glimpses on his personality.
He collaborated with Georgia-Tech, USA; NVIDIA, Dell-EMC, IBM, Amazon that resulted in faculty exchange, student exchange and research collaborations leading the department to make it a world class learning and research unit for Computing skills, innovation and knowledge creation. Submitted projects of around 50 million INR values to different agencies.
Excellence in driving the development & implementation of objectives for Industry Scale Consultancy and implementation of projects, curriculum and instructional evaluation, hiring quality faculty and generating International Grants, establishing Industrial and Foreign collaborations. Impeccable record of delivering over 200 talks, organizing international workshops on futuristic skills related to Machine Reinforcement Learning and Data Science in collaboration with professional organizations as well as workshops on Leadership for Indian Engineering institutions. Certified in Intelligent Video Analytics by NVIDIA.
Abstract:
AI is going through rapid transformation at unbelievable pace. Challenges to AI remain in terms of high end research in the area of Explainable AI, Responsible AI, Data biases and human in the loop. AI is also bringing tremendous changes in terms of work efficiencies and acting as an intelligent assistant for many tasks. Generative AI has further changed the whole landscape of innovations. Keynote will cover current status of AI on these lines.
Prof. Dr. Subarna Shakya
Prof. Dr. Subarna Shakya holds Ph.D. in Computer Engineering from Lviv Polytechnic National
University, Ukraine. He is Professor of Computer Engineering at Department of Electronics and
Computer Engineering, Pulchowk Campus, Institute of Engineering, Tribhuvan University and
also served as Visiting Professor in Brown University, Rhode Island, USA. He is also director of
IT Innovation Center, Tribhuvan University. He served as Executive Director at National
Information technology Center, Government of Nepal and also head of Department of
Electronics and Computer Engineering, Director of Center for Information Technology and
Chairman of Electronics and Computer Engineering Subject Committee, Institute of
Engineering, Tribhuvan University. He has also served as coordinator of EURECA (European
Research and educational collaboration with Asia) IDEAS (Innovation and Design for
EuroAsian Scholars) and LEADER (Links in Europe and Asia for engineering, education,
Enterprise and Research exchanges). Project is financed by the European Commission through
the Erasmus Mundus Program. He is the advisor member of National Information Technology
Committee, Government of Nepal. He has delivered over 30 Keynotes and invited speeches at
international conferences and workshops. He has published over 150 scientific/technical articles
and 5 books. He has been serving as an Editor/Guest Editor for over 15 international journals. He
is the expert member of Board of studies in South Asian University, India.
He is the Life Member of Indian society for mathematical modeling and Computer Simulation,
IIT, Kanpur, India, Senior Member of IEEE, member of the Society of Digital Information and
Wireless Communications, Senior Member of International Association of Computer Science
and Information Technology, fellow member of Scientific Society of Advanced research and
social change and Senior member of science and engineering institute, (SCIEI), www.sciei.org.
He has supervised more than 100 Masters Thesis, 10 Ph.D. thesis and 7 Ph.D. thesis are in
progress.
He was awarded by Nepal Education Leadership awards 2017, 18 Dec 2017 and outstanding
contribution to education, 17 Dec 2018 by World CSR Day and World Sustainability. He was
awarded 100 most dedicated professors, 4 th July, 2019 and also awarded best professor in
Computer Engineering studies, 10 th Dec 2019 by World education congress. He is Chief Editor of
journal of artificial Intelligence and capsule Networks (AICN). He has served as Chairman,
technical committee chairman and committee member in many International conferences such as
Springer and IEEE related to Computer Science and ICT as chairman. He is keen interest in
research and development of ICT, e-government system, Information security for e-Government
system, multimedia system, Computer Systems simulation and modeling, Cloud computing &
Security, Software & Information system and computer architecture.