본문 바로가기

K-DS

K-DS 콜로퀴움

강연안내
홈으로
강연안내
K-DS Distinguished Lecture Series

Making Discoveries for Humanity and Society
with Data Science

  • 강연 일시

    2024년 3월 29일 (금) 오후 2시

  • 연사 이름

    Meeyoung Cha

  • 소속

    Professor, KAIST Chief Investigator, IBS Scientific Director, MPI-SP

  • 문의

    K-DS 융합인재양성사업단 문의하기

연사소개

Meeyoung Cha

(Current) Professor at School of Computing, KAIST

(Current) Chief Investigator at the Institute for Basic Science (IBS)

(Current) Director, at Max Planck Institute for Security and Privacy (MPI-SP), Germany

(Former) Visiting Professor, Facebook Data Science Team, USA

Received ACM IMC Test-of-Time Award, AAAI ICWSM Test-of-Time Award, MSIT Award

강연개요

Making Discoveries for Humanity and Society with Data Science

Data science is an interdisciplinary field of study that extracts knowledge and insight from diverse forms of data through the use of scientific methods, algorithms, and systems. In particular, data science based on mathematical modeling and artificial intelligence (AI) has become critical for effectively handling astronomically growing data in various fields.

In this talk, I will introduce two research themes—poverty mapping and fake news detection—that address global issues and have a social impact. These research themes are examples of computational social science.

I will also talk about life as a data scientist, based on my experiences collaborating with world-class scientists at Facebook, AT&T Research, and Max Planck Institute, as well as NGOs like the United Nations Pulse Lab and the World Customs Organization.
K-DS Distinguished Lecture Series

How to Manage Societies Better than Optimal?

  • 강연 일시

    2024년 4월 26일 (금) 오후 4시

  • 연사 이름

    Dirk Helbing

  • 소속

    Professor of Computational Social Science, ETH Zurich

  • 문의

    K-DS 융합인재양성사업단 문의하기

연사소개

Dirk Helbing

Professor of Computational Social Science at ETH Zurich

More than 10 Publications in Nature, Science, and PNAS

Recipient of Honorary Ph.D. from TU Delft

Former Director of the Ph.D. School in "Engineering Social Technologies for a Responsible Digital Future" at TU Delft

Active Member of the External Faculty of the Complexity Science Hub Vienna

강연개요

How to Manage Societies Better than Optimal?

Given the ongoing digital revolution and our present-day sustainability challenges, the ways cities and societies are operated are currently being reinvented. Societies are complex systems. This has important implications for how a society should be managed: Not like a company, and also not like a machine! I will illustrate our insights by means of pattern formation and self-organisation phenomena in social systems as well as applications to self-governance and self-control. Based on this, I will argue that the requirement of organizing societies in a more resilient way implies the need for more decentralized solutions based on digitally assisted self-organization – a concept, which is also compatible with sustainability requirements and greater participation. I will further discuss, how collective intelligence and co-creation can be supported in ways that promote favourable systemic outcomes – outcomes that are better than optimal, i.e. better than when optimization is applied. As an application example, I will present a field study on participatory budgeting, which has been recently carried out in Aarau, Switzerland. Specifically, I will show, how voting rules can be improved to promote individual and systemic benefits, such as inclusion and fairness.
K-DS Distinguished Lecture Series

Human Mobility Science

  • 강연 일시

    2024년 5월 24일 (금) 오전 10시

  • 연사 이름

    Gautam Malviya Thakur

  • 소속

    Group Leader, Location Intelligence R&D, Oak Ridge National Laboratory

  • 문의

    K-DS 융합인재양성사업단 문의하기

연사소개

Gautam Malviya Thakur

(Current) Senior Staff Scientist and founding group leader of the Location Intelligence Group in the Geospatial Science and Human Security Division.

(Former) Deutsche Telekom Research Laboratories, Berlin, on transportation system modeling and understanding the network anatomy of major cities worldwide

(Former) Disney Research Laboratory, Zürich, on activity-driven mobility modeling of guests visiting the Disney theme parks

Research Interests in interconnected topics related to activity-driven human mobility modeling, place-based characterization, multi-scale global land use modeling, passive sensing, and spatially explicit disinformation detection.

Senior member of both ACM and IEEE

강연개요

Human Mobility Science

Human Mobility Modeling enables the understanding and characterization of time-variant movement and place visitation of individuals (or groups) in the real world. This knowledge is critical for domains like transportation planning, epidemic modeling, and energy use forecasting. Traditionally, simplistic statistical models such as a random walk or power law distribution were used to model these mobility patterns; however, it was determined that real-world mobility patterns are more complex and, therefore, require data-informed approaches for an accurate depiction. This talk will focus on the art and science of human mobility modeling with an emphasis on the use of Foundation Geoint data (demographics, points of interest, building footprint, etc.), behavior-driven movement-based patterns life generation, design and construction of data-informed human mobility model (HumoNet), spatio-temporal mobility interventions, and approaches to accurately benchmark mobility data to evaluate the presence of real-world mobility kinematics. The talk will conclude with the characterization of human mobility patterns from real-world scenarios, including disaster mitigation and cyber security attacks on critical infrastructure systems.
K-DS Distinguished Lecture Series

Frontiers of Collective Intelligence

  • 강연 일시

    2024년 6월 7일 (금) 오후 4시

  • 연사 이름

    Manuel Cebrián Ramos

  • 소속

    Senior Research Scientist, Spanish National Research Council

  • 문의

    K-DS 융합인재양성사업단 문의하기

연사소개

Manuel Cebrián Ramos

Current Position: Senior Research Scientist at the Center for Automation and Robotics, Spanish National Research Council; Member of the National Advisory Board for Artificial Intelligence, Spanish Government

Research Focus: Computational Social Science, Network Science, Artificial Intelligence

Past Affiliations: Max Planck Society, Massachusetts Institute of Technology (MIT), Commonwealth Scientific and Industrial Research Organisation (CSIRO), University of California at San Diego, Brown University

Publications: Articles in Science, Nature, and Proceedings of the National Academy of Sciences on computational approaches to societal challenges

Recognition: Research featured in The New York Times, The Economist, The Guardian

강연개요

Frontiers of Collective Intelligence

In today’s world, complex systems capable of solving problems under significant pressure are omnipresent, including social networks, cities, and the Internet. The advent of Artificial Intelligence (AI) has added a new dimension to these systems, igniting debates about its deployment and long-term societal impacts. This seminar diverges from the conventional AI discourse, spotlighting an alternative form of experimental intelligence: Collective Intelligence. Collective Intelligence, which has evolved alongside the Internet, is distinguished by the collective capacity to address challenges and make decisions in ways that surpass the capabilities of individual members. In this seminar, I will share insights from several pivotal situations I have encountered, which illustrate the application of Collective Intelligence. In this seminar, we will explore the versatile applications of Collective Intelligence, including locating hidden entities across vast geographies, coordinating global searches for missing individuals, reassembling critically important shredded documents, and directing human actors in real-time via the Internet. By examining both the untapped potential and the inherent challenges of Collective Intelligence, the seminar aims to foster a dialogue on how the lessons learned from these experiences can inform the future development and strategic deployment of Artificial Intelligence.
K-DS Distinguished Lecture Series

Generative AI meets Vector databases and RAG techniques

  • 강연 일시

    2024년 9월 20일 (금) 오후 2시

  • 연사 이름

    Dr. Min-Soo Kim

  • 소속

    Professor at School of Computing, KAIST

  • 문의

    K-DS 융합인재양성사업단 문의하기

연사소개

Dr. Min-Soo Kim

(Current) Professor at School of Computing, KAIST

(Current) Co-Founder and CEO of GraphAI Co. Ltd.

(Current) Vice President, Office of Academic Information, KAIST

(Former) Director of Research Center for Extreme Exploitation of Dark Data, ERC, NRF, Korea

(Former) Professor at Electrical Engineering and Computer Science, DGIST

(Former) Postdoctoral Research Staff, IBM Almaden Research Center, USA

강연개요

Generative AI meets Vector databases and RAG techniques

In numerous industrial sectors, significant efforts are underway to enhance efficiency and automate operations utilizing generative AI models such as ChatGPT. Despite their potential, these technologies encounter three primary challenges: hallucination, data recency, and high cost. In response, the Retrieval-Augmented Generation (RAG) technique has emerged as a promising solution, receiving considerable attention for its potential to mitigate these challenges. RAG addresses hallucination and data recency problems by enabling generative AI models to access and retrieve relevant facts from databases, subsequently generating responses based on this retrieved data. Vector databases, which are extensively employed for this purpose, embed diverse data types, including documents, images, and videos, into multi-dimensional vectors. To effectively counter the curse of dimensionality, vector databases typically employ graph-based indexes. RAG has been enhanced with variations like iterative RAG and plan RAG to tackle more complex challenges. Iterative RAG repeatedly executes the retrieval and generation cycle, while plan RAG strategizes to solve intricate corporate decision-making problems prior to retrieval and generation. In this talk, I will introduce these innovative methodologies about vector databases and RAG techniques.
K-DS Distinguished Lecture Series

Query optimization in cloud database, and Graph neural networks: techniques and benchmarks

  • 강연 일시

    2024년 10월 25일 (금) 오전 10시

  • 연사 이름

    Makoto Onizuka

  • 소속

    Professor, Graduate School of Information Science and Technology, Osaka University

  • 문의

    K-DS 융합인재양성사업단 문의하기

연사소개

Makoto Onizuka

Makoto Onizuka is a Professor at Graduate School of Information Science and Technology, Osaka University. He is the leader of Big data engineering Laboratory and conducts research on graph mining algorithms and AI-driven database query optimization techniques. Prior to joining Osaka University, he worked at Nippon Telegraph and Telephone Corporation (NTT) for more than 20 years being served as a distinguished technical member from 2010 to 2014. He also worked as a visiting scholar at the University of Washington from 2000 to 2001. He developed research prototype systems and some of them were used in production: LiteObject (object-relational main memory database system), XMLToolkit (XML stream engine), CBoC type2 (Common IT Bases over Cloud Computing at NTT), and Grapon (Graph mining techniques). He serves as Director at the database society of Japan (2024-present), Information Processing Society of Japan (2019-2021), Academic committee at Shonan meeting (2016-present), PC co-chair at DASFAA 2024 and KJDB 2024, Publicity co-chair at MIPR 2021, Workshop co-chair at VLDB 2020, Best demonstration award committee at VLDB 2024 and DASFAA 2010, Best paper award committee at DASFAA 2012, and program committee at international conferences, including VLDB(2009-2010 industrial track, 2024 demonstration track), SIGMOD(2018, 2020), ICDE(2015 industrial track, 2023 demonstration track), AAAI(2021-2025), NeurIPS(2022-2025), ICLR(2024), IJCAI(2023-2024), ICML(2023-2024), ECML(2022, 2024), CIKM(2017-2018,2020-2023), DASFAA(2010-2016,2020-2023).

강연개요

Query optimization in cloud database, and Graph neural networks: techniques and benchmarks

In this talk, I will introduce recent advancements in cloud database management and graph neural networks.

Regarding cloud database management, many companies are striving to manage their data in the cloud and are working on improving sales and cost reduction through real-time behavioral analysis. I will discuss technological trends in this area and our efforts, including cloud-scale database testing, schema optimization problem for time-dependent workloads, and join query optimization using cardinality estimation.

Next, in the context of graph neural networks (GNNs), we will address the explosive growth of research papers in this field, driven by recent advances in AI. We will introduce trends in GNNs and our efforts on improving the performance of GNNs, GNN benchmarking by synthetic graph generators, and patent search using GNNs.
K-DS Distinguished Lecture Series

Privacy in the Age of AI and Large Language Models

  • 강연 일시

    2024년 11월 22일 (금) 오전 10시

  • 연사 이름

    Professor Li Xiong

  • 소속

    Samuel Candler Professor of Computer Science and Biomedical Informatics, Emory University

  • 문의

    K-DS 융합인재양성사업단 문의하기

연사소개

Professor Li XIONG

Li Xiong is a Samuel Candler Professor of Computer Science and Biomedical Informatics at Emory University. She has a Ph.D. from Georgia Institute of Technology, an MS from Johns Hopkins University, and a BS from the University of Science and Technology of China. Her research lab, Assured Information Management and Sharing (AIMS), conducts research on trustworthy and privacy-enhancing data-driven AI solutions for healthcare, public health, and spatial intelligence. She is recognized as an IEEE fellow for her contributions on privacy-preserving and secure data sharing and analytics. She has published over 200 papers and received six best paper or runner up awards. She has served and serves as associate editor for TKDE, TDSC, VLDBJ, general or program chair for SIGSPATIAL, CIKM, BigData, and program vice chair for SIGMOD, VLDB, and ICDE. Her research has been supported by both governments (NSF, NIH, IARPA, AFOSR) and industry/foundations (Mistubishi, Cisco, AT&T, Google, IBM). More details are at http://www. cs.emory.edu/~lxiong.

강연개요

Privacy in the Age of AI and Large Language Models

As artificial intelligence (AI) and large language models (LLMs) increasingly influence every facet of our lives, ensuring the privacy of user data has become paramount. In this talk, I will review the commonly used technique for ensuring differential privacy of machine learning models built from privacy sensitive data and introduce several of our recent works on:
1) ensuring differential privacy for training graph neural networks from correlated graph data,
2) ensuring personalized differential privacy for federated learning from decentralized data, and
3) new privacy risks and defenses in the emerging fine-tuning paradigm using pre-trained LLMs.
K-DS Distinguished Lecture Series

Under the Ground, Over the Oceans, and through the Space: Exploring the World with Imaging Beyond the Visible Spectrum.

  • 강연 일시

    2024년 12월 13일 (금) 오후 4시

  • 연사 이름

    Professor Seniha Esen Yuksel

  • 소속

    Department of Electrical and Electronics Engineering, Hacettepe University

  • 문의

    K-DS 융합인재양성사업단 문의하기

연사소개

Professor Seniha Esen Yuksel

Dr. Seniha Esen Yuksel received her Ph.D. degree from the University of Florida, Department of Computer Information, Science and Engineering, USA in 2011. Currently, she is an associate professor at Hacettepe University, Department of Electrical and Electronics Engineering. She also serves as the director of the Pattern Recognition and Remote Sensing Laboratory (PARRSLAB), where her research focuses on machine learning and computer vision utilizing sensors that extend beyond the visible spectrum, such as hyperspectral, radar, X-ray, thermal, SAR, and LiDAR. Dr. Yuksel is recognized as an IEEE Senior member and has been honored with the BAGEP Outstanding Young Scientist Award by the Science Academy. She has contributed to the scientific community by authoring or coauthoring over 100 articles published in peer-reviewed journals and conferences.

강연개요

Under the Ground, Over the Oceans, and through the Space: Exploring the World with Imaging Beyond the Visible Spectrum.

Hyperspectral cameras are specialized devices that capture hundreds of images from across the electromagnetic spectrum. This comprehensive collection allows for the identification and labeling of each pixel based on its material composition, be it grass, water, gold, cotton, and more. The wealth of information provided by hyperspectral imaging has far-reaching applications in diverse fields such as agriculture, astronomy, medical imaging, and defense. In this presentation, I will introduce our research which ventures beyond the limitations of the visible spectrum using radar and hyperspectral images, seeing under the ground, over the oceans and through space. Additionally, I will outline our current research on deep denoising networks for hyperspectral imaging (HSI), introduce the current trends, discuss the future of HSI with the current paradigm shifts in the imaging pipeline, and touch upon the low-cost programmable cameras that put the HSI capability to our cell-phones.