IEEE DSAA 2026

Abstract: We are living in an era where our digital lives are increasingly interdependent and deeply interconnected. These connections rely on a vast, layered ecosystem of actors—many of whose trustworthiness is uncertain or outright suspect. Over the past three decades, rapid advances in computing and communication technologies have brought unprecedented access and connectivity to billions of users. Yet this digitization comes at a cost: our interactions, queries, and data are increasingly vulnerable to privacy violations. Today, threats to privacy come not just from malicious individuals, but also from powerful institutions—ranging from service providers to nation-states. In this reality of an untrusted world, we pose several foundational research questions: (i) Can we design a scalable voice communication system that ensures absolute privacy? (ii) Can we build an oblivious search engine over public document repositories? (iii) Can we develop scalable private query processing over shared or public databases? (iv) And in the age of large language models, can we enable private inference for user queries? These are not just open problems — they are essential challenges if we are to build trusted services over untrusted infrastructures. In this talk, I will present recent work that leverages Homomorphic Encryption to address some of these questions. We explore the inherent performance and scalability trade-offs in enabling private access, search, and inference. If nothing else, our results underscore a critical insight: ensuring privacy at scale is not impossible, but it comes at a high cost.

Biography: Divy Agrawal is a Distinguished Professor and Chair of Computer Science at the University of California, Santa Barbara (UCSB), where he also holds the Leadership Endowed Chair in the Department of Computer Science. He received his B.E. (Hons.) in Electrical Engineering from BITS Pilani, followed by M.S. and Ph.D. degrees in Computer Science from the State University of New York at Stony Brook. Since joining UCSB, Professor Agrawal has established himself as a leading researcher in databases, distributed systems, cloud computing, and large-scale data infrastructures and analytics. Over the course of his career, he has published more than 400 research articles and mentored approximately 50 Ph.D. students. He currently serves as Editor-in-Chief of both the Proceedings of the ACM on Modeling of Data and the Springer journal Distributed and Parallel Databases. He has served on several editorial boards, including ACM Transactions on Database Systems, IEEE Transactions on Knowledge and Data Engineering, ACM Transactions on Spatial Algorithms and Systems, ACM Books, and the VLDB Journal. Professor Agrawal is a former Trustee of the VLDB Endowment and recently served as Chair of the ACM Special Interest Group on Management of Data (SIGMOD). His recognitions include the Gold Medal from BITS Pilani, the UCSB Academic Senate Award for Outstanding Graduate Mentoring, and multiple paper honors: Best Paper Awards (ICDE 2002, MDM 2011), an Influential Paper Award (NDSS 2024), and Test-of-Time Awards (ICDT, MDM). He is a Fellow of the ACM, IEEE, and AAAS.

Abstract: Integrating causality into machine learning provides a principled foundation for achieving robustness under distribution shifts and hallucinations and provides a framework for explainable AI. Causality ensures stability across diverse operational conditions and allows models to learn to respond to interventions and shifting environments. In our talk, we shall show how domain-based features can be exploited to ensure robust and explainable AI performance in different domains. We shall be considering two contrasting application domains – medical imaging and campaign designs for commercial interventions.

Biography: Professor Santanu Chaudhury is the first Dean of the Vachani School of Advanced Computing at Ashoka University. Prof. Santanu Chaudhury did his B.Tech (1984) in Electronics and Electrical Communication Engineering and PhD (1989) in Computer Science and Engineering from IIT Kharagpur, India. He has earlier been Director, IIT Jodhpur and Director CSIR-Central Electronics Engineering Research Institute. He was also Dean, Undergraduate Studies at IIT Delhi and has held a number of Chair Professor positions at IIT Delhi. He was awarded the INSA medal for young scientists in 1993. He is a fellow of the Indian National Academy of Engineers (INAE), the National Academy of Sciences (NASI) and the International Association of Pattern Recognition (IAPR). He is also a recipient of the Distinguished Alumni Award of IIT Kharagpur.He has over 350 publications in reputed Journals and conferences and 15 patents with technologies commercialised by global industries. He has authored and edited books on Multimedia Ontology and Digital Heritage. His areas of interest are Computer Vision, Artificial Intelligence, AI Applications, Digital Heritage, AR-VR & Multi-sensory media. He has led several national initiatives across the fields of robotics, medical imaging, cyber-physical systems, document image understanding, and digital heritage and has been instrumental in advancing interdisciplinary AI research in the country.

Abstract: The current Generative AI (GAI) revolution is built on a simple premise: bigger is better. Tech giants are fiercely chasing the 'scaling law,' betting that massive data and immense compute will unlock Artificial General Intelligence (AGI). This talk challenges that assumption, arguing that scaling alone is insufficient to achieve AGI. Grounded in insights from social media mining and data science, we propose Interdisciplinary Intelligence, a collaborative framework essential for the next frontier of AI. This framework not only drives technical innovation but also tackles the urgent socio-technical challenges of the LLM era, including bias, reliability, equity, evaluation, and the integrity of AI-assisted research. This talk serves as a call to action for computer scientists, data scientists, social scientists, ethicists, and policymakers to come together and co-create the future of responsible AI.

Biography: Prof. Huan Liu is a Regents professor of computer science and engineering with the School of Computing and Augmented Intelligence in the Ira A. Fulton Schools of Engineering. At Arizona State University, he was recognized for excellence in teaching and research in Computer Science and Engineering and received the 2014 President's Award for Innovation. He is the recipient of the ACM SIGKDD 2022 Innovation Award. His research interests are in data mining, machine learning, feature selection, social computing, and artificial intelligence. He is a co-author of a text, Social Media Mining: An Introduction, Cambridge University Press. He is a founding organizer of the International Conference Series on Social Computing, Behavioral-Cultural Modeling, and Prediction, and Chief Editor of Data Mining and Management in Frontiers in Big Data. He is a Fellow of ACM, AAAI, AAAS, and IEEE.

Abstract: In this talk, we describe a neural-symbolic architecture for explaining rare events. We propose a two-layer system, where the first layer is unsupervised, based on autoencoders, and is used designed to detect rare events (outliers, drifts, anomalies, failures). The second layer is supervised, based on rule learners, and is used to explain the anomalies detected in the first layer. Both models run online and in parallel. We evaluate the proposed system in a real-world case study of predictive maintenance. We present examples of explanations that illustrate their benefits.

Biography: João Gama is an Emeritus Professor at the University of Porto, Portugal. He received his Ph.D. in Computer Science from the University of Porto in 2000. He taught Informatics and data sciences at the School of Economics for more than 30 years. He is EurAI Fellow, IEEE Fellow, and Fellow of the Asia-Pacific AI Association. He is a member of the Academia das Ciências de Lisboa. His main scientific contributions are in the area of learning from data streams, where he has an extensive list of publications. He is the Editor-in-Chief of the International Journal of Data Science and Analytics, published by Springer.