Keynote SPEECH FRI 29 JULY – PROFESSOR GEORGE KARYPIS
title: Graph Neural Network Research at AWS AI
Abstract: In the course of just a few years, Graph Neural Networks (GNNs) have emerged as the prominent supervised learning approach that brings the power of deep representation learning to graph and relational data. An ever-growing body of research has shown that GNNs achieve state-of-the-art performance for problems such as link prediction, fraud detection, target-ligand binding activity prediction, knowledge-graph completion, and product recommendations. As a result, GNNs are quickly moving from the realm of academic research involving small graphs to powering commercial applications and very large graphs. This talk will provide an overview of some of the research that AWS AI has been doing to facilitate this transition, which includes developing the Deep Graph Library (DGL)—an open-source framework for writing and training GNN-based models, improving the computational efficiency and scaling of GNN model training for extremely large graphs, developing novel GNN-based solutions for different applications, and making it easy for developers to train and use GNN models by integrating graph-based ML techniques in graph databases.
Bio: Professor George Karypis is a Senior Principal Scientist at AWS AI and a Distinguished McKnight University Professor and an ADC Chair of Digital Technology at the Department of Computer Science & Engineering at the University of Minnesota. He is a Fellow of the IEEE. His research interests span the areas of data mining, machine learning, high performance computing, information retrieval, collaborative filtering, bioinformatics, cheminformatics, and scientific computing. His research has resulted in the development of software libraries for serial and parallel graph partitioning (METIS and ParMETIS), hypergraph partitioning (hMETIS), for parallel Cholesky factorization (PSPASES), for collaborative filtering-based recommendation algorithms (SUGGEST), clustering high dimensional datasets (CLUTO), finding frequent patterns in diverse datasets (PAFI), and for protein secondary structure prediction (YASSPP). He has coauthored over 300 papers on these topics and two books (“Introduction to Protein Structure Prediction: Methods and Algorithms” (Wiley, 2010) and “Introduction to Parallel Computing” (Publ. Addison Wesley, 2003, second edition)). In addition, he is serving on the program committees of many conferences and workshops on these topics, and on the editorial boards of the IEEE Transactions on Knowledge and Data Engineering, ACM Transactions on Knowledge Discovery from Data, Data Mining and Knowledge Discovery, Social Network Analysis and Data Mining Journal, International Journal of Data Mining and Bioinformatics, the journal on Current Proteomics, Advances in Bioinformatics, and Biomedicine and Biotechnology.
KEYNOTE SPEECH SAT 30 JULY – PROFESSOR MILOS MANIC
TITLE: EXPLAINABLE AI IN Cyber-Physical Systems
Abstract: Cyber-Physical Systems (CPS) play a critical role in our modern infrastructure due to their capability to connect computing resources with physical systems. As such, topics such as reliability, performance, and security of CPSs continue to receive increased attention from the research community. CPSs produce massive amounts of data, creating opportunities to leverage Artificial Intelligence (AI) for performance monitoring and optimization, preventive maintenance, and threat detection. However, the opaque nature of complex AI models is a drawback when used in safety-critical systems such as CPS. While explainable AI (XAI) has been an active research area in recent years, much of the work has been focused on supervised learning. As CPS rapidly produce massive amounts of unlabeled data, relying on supervised learning alone is not sufficient for data-driven decision making in CPS. This talk will outline how unsupervised explainable AI could be used within CPS, including a review of the state-of-the-art in XAI, present initial desiderata of explainable AI for CPS, followed by novel unsupervised learning approaches for XAI that generate global and local explanations.
Bio: Milos Manic is a Professor with the Computer Science Department at Virginia Commonwealth University and is the director of the VCU Cybersecurity Center. He is also a Commonwealth Cyber Initiative Fellow, inaugural class 2020-2022. As a principal investigator or university partner, he has completed more than 40 research grants with the departments of Energy, Homeland Security, Air Force, Battelle Energy Alliance/Idaho National Laboratory, National Science Foundation, and industry entities, in the area of data mining and machine learning applied to cybersecurity, critical infrastructure protection, energy security, and resilient intelligent control. Manic has given over 40 invited talks around the world, authored more than 200 refereed articles in international journals, books, and conferences, holds several U.S. patents and won the 2018 R&D 100 Award for Autonomic Intelligent Cyber Sensor (AICS), one of top 100 science and technology worldwide innovations in 2018. He is also an inductee of U.S. National Academy of Inventors (class of 2019). He is an IEEE Fellow, recipient of IEEE IES 2019 Anthony J. Hornfeck Service Award. He also received the 2012 J. David Irwin Early Career Award and 2017 IEM Best Paper Award.
KEYNOTE SPEECH SUN 31 JULY – PROFESSOR DAMMINDA ALAHAKOON
TITLE: TOWARDS AN Artificial GENERAL Intelligence WITH Self-Structuring AI
Abstract: Self-Structuring Artificial Intelligence (SS-AI) is a new paradigm of Artificial Intelligence that has demonstrated potential of building an Artificial General Intelligence. SS-AI broadly aligns to the domain of unsupervised machine learning, and can be defined as learning structures that autonomously evolve with the unstructured and unlabelled nature of data; spatially, temporally, laterally and semantically. This innovative approach towards a new type of artificial intelligence is based on theories of cognitive psychology, computational neuroscience, affective computing and self-organisation. This talk will present the concept, design and development of new SS-AI algorithms that leverage the cognitive feature representation of vector symbolic architectures (VSA) to learn a topological formulation of the base vectors of an unlabelled data space in just a few updates. It follows a unique ‘loser-takes-all approach for neuronal excitations, in contrast to winner-take-all as practiced in most conventional neural network learning methods.
Bio: Professor Damminda Alahakoon is the Founding Director of the Centre for Data Analytics and Cognition (CDAC) at La Trobe University, Australia. Damminda has made
significant contributions with international impact towards the advancement of Artificial Intelligence through academic research, applied research, research supervision, industry engagement, curriculum development and teaching. He has published over 100 research articles; theoretical research in self-structuring AI, human-centric AI, cognitive computing, deep learning, optimization, and applied AI research in industrial informatics, smart cities, robotics, intelligent transport, digital health, energy, sport science and education. He is the primary investigator of research grants totaling more than $10 million from the public and private sector.