Abstract: Decentralized AI requires collaborations among untrusted entities while maintaining privacy, identity-management and security. Can Web3 play a role in Decentralized AI? Web3 has ushered rapid innovation in the financial sector and the transaction layer. What about other societal sectors and other layers of their stack? Can a fusion of Web3 and AI allow a wider adoption? The elements of web3: incentives, ledger-tech and pseudo-anonymity will be impactful in many sectors that involve societal coordination such as health, transportation, e-commerce, civic service and more. This involves providing decentralization of more than just the transaction layer and the related meta-data. We need to support many more data-centric solutions such as machine learning, decision guidance, digital twins and more. Four new components are critical to upgrade Web3 technologies to support Decentralized AI : (1) distributed verifiable AI (beyond distributed ledger) to harness invisible data, (2) privacy-tech (beyond pseudo anonymity) to overcome data silos, (3) data markets (beyond game-theory of cryptocurrencies) to incentivize + facilitate sharing and (4) CrowdX:crowd experience (beyond crypto exchanges) for societal orchestrion. How will the privacy, security, web3 and AI community build these solutions? What are the big research problems? What are the adjacent innovations? And what are the risks? For more details on the MIT program, please see http://desoc.media.mit.edu/
Bio: Ramesh Raskar is an Associate Director and Associate Professor at MIT Media Lab and directs the Camera Culture research group. He codirects the MIT Program on Decentralized Society+Web3. His focus is on AI and Imaging for health and sustainability. They span research in physical (e.g., sensors, health-tech), digital (e.g., automated and privacy aware machine learning) and global (e.g., geomaps, autonomous mobility) domains. He received the Lemelson Award (2016), ACM SIGGRAPH Achievement Award (2017), DARPA Young Faculty Award (2009), Alfred P. Sloan Research Fellowship (2009), TR100 Award from MIT Technology Review (2004) and Global Indus Technovator Award (2003). He has worked on special research projects at Google [X], Facebook, Apple and co-founded/advised several companies. https://www.media.mit.edu/people/raskar/overview/
Abstract: With growing organizational infrastructure in terms of size and complexity, it becomes increasingly difficult to manually configure, manage and administer security. Such automation not only enables the correct assignment of user permissions, it also aids in identifying misconfigurations (over privileges or under privileges). While Role-Based Access Control (RBAC) is the widely deployed security policy as a means to enforce access control, Attribute Based Access Control (ABAC) is becoming an attractive choice due to its flexibility, scalability, dynamic nature, portability and identity-less nature. In this talk, we consider both RBAC and ABAC and discuss the benefits of automating the process of security policy configuration and administration, and discuss the research challenges and solutions including automatic discovery of policies for both RBAC and ABAC using mining, incrementally maintain the policy, that does not require remining, when changes occur, and automating the process of cross organizational policy translation, reconciliation and negotiation. While many of these problems are NP-hard, we discuss existing solutions that offer reasonable performance without sacrificing accuracy. We then identify open problems and challenges.
Bio: Dr. Atluri is a Professor of Computer Information Systems in the MSIS Department, and director of the Center for Information Management, Integration and Connectivity (CIMIC) at Rutgers University. Dr. Atluri’s research interests include Information Security, Privacy, Databases, Workflow Management, Spatial Databases and Distributed Systems and has pioneering research in workflow security. She was the recipient of the National Science Foundation CAREER Award in 1996, Rutgers University Research Award for untenured faculty for outstanding research contributions in 1999, 2014 outstanding research award from the IFIP WG11.3 data and application security and privacy, 2022 ESORICS Outstanding Contribution Award, and Distinguished Alumni of College of Engineering, Kakinada, Nagarjuna University, India, in 2022. Her research has been sponsored by NSF, DHS, DoD, NSA, ARO, NOAA, EPA, Lawrence Livermore National Laboratory, Hackensack Meadowlands Development Commission and SAP Research.
Abstract: Data markets are emerging and promising for harvesting data from many data owners to support data-driven AI applications and many second-uses of big data. Data valuation, such as pricing, plays a central role in data markets. In this talk, I will survey the motivations and the state-of-the-art practice of data and (machine learning) model markets, and review data valuation in end-to-end data analytics and machine learning pipelines. Then, I will focus on models, fairness, and scalability of data valuation using some well established solution concepts in cooperative game theory, such as Shapley value. As a principled approach, I will illustrate that with some simple yet practical assumptions about the utility of data products, assessing accurate Shapley value of millions of products and tens of owners is highly practical. I will also demonstrate the challenges in modeling and computing fair reward allocation in one-shot cooperative machine learning processes, such as federated learning, as well as in building privacy preserving model marketplaces.
Bio: Jian Pei is Professor at Duke University. His research focuses on data science, data mining, database systems, information retrieval and applied machine learning. His expertise is on developing effective and efficient data analysis techniques for novel data intensive applications and transferring to products and business practice. He is recognized as a Fellow of the Royal Society of Canada (Canada's national academy), the Canadian Academy of Engineering, ACM and IEEE. He received several prestigious awards, including the 2017 ACM SIGKDD Innovation Award, the 2015 ACM SIGKDD Service Award, and the 2014 IEEE ICDM Research Contributions Award. He was a past chair of ACM SIGKDD and a past EIC of IEEE TKDE.
Abstract: The representation of women and underrepresented minority communities has increased a great deal in Computer Science. However, in many disciplines of Computer Science such as Data Science and Cyber Security, it is vastly underrepresented. Far fewer women have been elected to IEEE and ACM Fellows than men when it should be around 50%. In addition, fewer women are in positions of power both in academic institutions (e.g., Engineering Deans) and in the C-Suite in corporations as well as on corporate boards. This is partly because of the number of women at say first level management are far fewer than men. Then you have to rise up the ladder from that pool and so women are already at a disadvantage. As a result, we do not have a sufficient number of women who are senior researchers and practitioners to support younger women in getting promotions and awards such as IEEE and ACM Fellows and various Technical Recognition Awards. As a result, women not only lose out on the recognitions, but they also lose out on high paying jobs. One way for women and underrepresented minority communities (or for that matter any community) to have an edge is to not only do a BS and MS in Computer Science (CS) but also a PhD in CS. While it takes around three more years to complete PhD after a two year MS degree, a PhD will open many doors in different sectors including in academia, industry, and government, but also it enables one to obtain high paying jobs. A high paying job is a must especially for every woman. In addition, it is also critical that women and those from the underrepresented communities to get mentors to give them advice, speak for them when needed, and also provide them with opportunities to advance in their careers. Women and underrepresented groups must understand the culture of an organization to succeed. In addition, they must also form support groups. We are living in a complex world that is rapidly evolving due to technology. While there are numerous career opportunities in areas like Data Science and Cyber Security, the competition is also extremely intense around the globe. As the population gets more educated, the challenge is how do you differentiate yourself from the others? While working hard is necessary, it is not sufficient. In this presentation I will provide a brief background on how I got here and how I benefitted by doing a PhD. I will then discuss the benefits of getting into careers in areas such as Cyber Security and Data Science, the benefits of doing a PhD in Computer Science, and other aspects such as getting mentors and forming support groups. I will also give examples from my own experience.
Bio: Dr. Bhavani Thuraisingham is the Founders Chair Professor of Computer Science, the Founding Executive Director of the Cyber Security Research and Education Institute (2004-2021), and the Co-Director of the Women in Cyber Security and Women in Data Science Centers at the University of Texas at Dallas. She is also a visiting senior research fellow at Kings College, the University of London since 2015 and gives lectures pro-bono in Trustworthy Machine Learning at the University of Dschang, Cameroon Africa. Dr. Bhavani is an elected Fellow of several prestigious organizations including the ACM, the IEEE, the AAAS, and the NAI (National Academy of Inventors). Her research, development and education efforts have been on integrating cyber security and data science/machine learning for the past 37 years including at Honeywell Inc., The MITRE Corporation, the National Science Foundation, and Academia. Her recent focus has been on developing scalable trustworthy machine learning-based solutions for tackling the Cyber Security challenges as well as on various aspects of Cyber Governance and Risk. She has received several awards including the IEEE Computer Society’s 1997 Technical Achievement Award, ACM SIGSAC 2010 Outstanding Contributions Award, 2013 IBM Faculty Award, 2017 ACM CODASPY (Data and Applications Security and Privacy) Lasting Research Award, the 2017 Dallas Business journal Women in Technology Award, and the 2019 IEEE ComSoc Technical Recognition Award for Communications and Information Security. She has delivered over 200 keynote/featured addresses and over 120 panel presentations including for Fortune Media, Dell Technologies World, Professors Without Borders, and the renowned Lloyd’s of London Insurance, written 16 books, published over 130 journal articles and over 300 conference papers and has 7 US patents. She has also written opinion columns on security for venues such as the New York Times, WomensDay.com, Inc. Magazine and the Legal 500. Dr. Thuraisingham received her PhD in Computability Theory from the University Wales, UK and the prestigious earned higher doctorate (D.Eng) form the University of Bristol, England for her published work in Secure Data Management. She also has a certificate in Public Policy Analysis from the London School of Economics.
Abstract: Software is increasingly playing a key role in all infrastructure and application domains we may think of. Unfortunately, as we all know, software systems are still often insecure, despite the fact the “problem of software security” had been known to the industry and research communities for decades. In this talk, as an example of insecure software, we present the results of an extensive study to detect vulnerable implementations of pseudo-random number generator (PRNG) in mobile apps. The study has been carried out using an analysis tool, OTP-Lint that assesses implementations of the PRNGs in an automated manner without requiring the source code. By analyzing 6,431 commercial apps downloaded from two well-known apps market, OTP-Lint identified 399 vulnerable apps that generate predictable OTP values. We then discuss other factors that today complicate the problem of software security - a notable factor being the software supply chain. We then discuss "what it takes" to convince all parties involved in the software ecosystem to address the problem of software insecurity and outline research directions.
Bio: Elisa Bertino is Samuel Conte professor of Computer Science at Purdue University. She serves as Director of the Purdue Cyberspace Security Lab (Cyber2Slab). Prior to joining Purdue, she was a professor and department head at the Department of Computer Science and Communication of the University of Milan. She has been a visiting researcher at the IBM Research Laboratory in San Jose (now Almaden), at Rutgers University, at Telcordia Technologies. She has also held visiting professor positions at the Singapore National University and the Singapore Management University. Her main research interests include security, privacy, database systems, distributed systems, and sensor networks. Her recent research focuses on cybersecurity and privacy of cellular networks and IoT systems, and on edge analytics for cybersecurity. Elisa Bertino is a Fellow member of IEEE, ACM, and AAAS. She received the 2002 IEEE Computer Society Technical Achievement Award for “For outstanding contributions to database systems and database security and advanced data management systems”, the 2005 IEEE Computer Society Tsutomu Kanai Award for “Pioneering and innovative research contributions to secure distributed systems”, the 2019-2020 ACM Athena Lecturer Award, and the 2021 IEEE 2021 Innovation in Societal Infrastructure Award.
Abstract: Chronic disease is a major epidemic in the 21st century, with more than 50% of the world population suffering some form of chronic disease leading to lowered quality of life and shorter lives. The vast majority of chronic diseases are due to environmental and lifestyle factors, having a major impact on the molecular and biological pathways. If we can understand these pathways, we can provide health insights and nutritional recommendations that are precisely tailored to individuals to maintain a healthy state. We can also detect disruptions in those pathways as early as possible to provide health interventions that can save lives. We have collected molecular data and clinical metadata for almost half a million samples from around the world, and extracted human and microbiome gene expression profiles. We have analyzed this high-dimensional and longitudinal data using AI and ML techniques, and developed a suite of clinical applications that have the potential to improve human health globally. In this talk, I will discuss the progress we have made and the remaining challenges.
Bio: Guru is the founding CTO of Viome Inc, where he leads AI, discovery, and engineering. In this role, he leads the development of Viome’s molecular data analysis platform, based on which a range of products deliver precision insights and interventions to deter chronic disease to hundreds of thousands of customers around the world. The core of this platform is based on extracting novel insights from large gene expression datasets (from stool, blood, saliva, and other samples), and from clinical studies targeting disorders including metabolic, gastrointestinal, and cancer. Guru recently led the development of a first-of-a-kind saliva-based early cancer detection system for oral and throat cancer, which won FDA’s designation as a breakthrough device. He previously led IBM Watson AI Research, to develop and commercialize advanced technology solutions in multiple industries. Guru has published and patented extensively, been featured in international media, and received various awards and recognitions for his accomplishments.