Osaka University Logo Graduate School of IST Logo

Winter School · Trustworthy & Generative AI for Society 5.0 (Online)

Graduate School of Information Science and Technology — Osaka University

Advancing Human‑Centered, Trustworthy, and Generative AI to Build a Resilient Society 5.0

Online sessions (English) with interactive Q&A. Typical session length: 1.5–2 hours (including Q&A). Zoom link is emailed after registration.

The Graduate School of Information Science and Technology (IST), The University of Osaka, is pleased to announce the IST International Webinar Series, an online lecture series presented in English by our faculty members. This series covers advanced topics in information science and technology, including AI, graph data analysis, distributed algorithms, wireless systems, and hardware security. Each session includes a live Q&A discussion. The detail is as follows.

Reserve Your Seat

Dates

March 10–13 · Online

Host

Graduate School of IST, Osaka University

Speakers

Assoc. Prof. Yoshihiro Toya
Talk 1 · Tue, Mar 10 · 09:00 JST

Engineering of Metabolic Pathways and Controlling the Flux for Efficient Bio‑Production

Assoc. Prof. Yoshihiro Toya · The University of Osaka

Assoc. Prof. Shunsuke Saruwatari
Talk 2 · Wed, Mar 11 · 02:00 JST

Millimeter-level Wireless Localization Using UWB Wireless Networks

Assoc. Prof. Shunsuke Saruwatari · The University of Osaka

Assoc. Prof. Yuya Sasaki
Talk 3 · Wed, Mar 11 · 17:00 JST

Graph Data Analysis: From Efficiency and Accuracy to Trustworthy

Assoc. Prof. Yuya Sasaki · The University of Osaka

Prof. Taisuke Izumi
Talk 4 · Thu, Mar 12 · 17:00 JST

Distributed Algorithm Beyond Worst Case

Prof. Taisuke Izumi · The University of Osaka

Assoc. Prof. Jun Shiomi
Talk 5 · Fri, Mar 13 · 17:00 JST

Towards a Safe and Secure Society: Detection of Hardware Trojans in Integrated Circuits

Assoc. Prof. Jun Shiomi · The University of Osaka

Schedule

A focused set of talks from March 10–13. All sessions are online via Zoom (JST). Links are emailed automatically after Google Form registration. Each session typically runs 1.5–2 hours including Q&A.

Tue, March 10 (JST)

09:00
Assoc. Prof. Yoshihiro Toya

Engineering of Metabolic Pathways and Controlling the Flux for Efficient Bio‑Production

Wed, March 11 (JST)

02:00
Assoc. Prof. Shunsuke Saruwatari

Millimeter-level Wireless Localization Using UWB Wireless Networks

17:00
Assoc. Prof. Yuya Sasaki

Graph Data Analysis: From Efficiency and Accuracy to Trustworthy

Thu, March 12 (JST)

17:00
Prof. Taisuke Izumi

Distributed Algorithm Beyond Worst Case

Fri, March 13 (JST)

17:00
Assoc. Prof. Jun Shiomi

Towards a Safe and Secure Society: Detection of Hardware Trojans in Integrated Circuits

Register

Participation is free but registration is required. Upon registration, attendees will receive the Zoom link by email. Please use the Google Form below to register and save your info.

Open Registration Form

Assoc. Prof. Yoshihiro Toya
Yoshihiro Toya

Engineering of Metabolic Pathways and Controlling the Flux for Efficient Bio‑Production

Time: Tue, Mar 10 & Fri, Mar 13, 09:00 JST

Abstract: To achieve a sustainable society, many efforts are being made to replace chemical processes that depend on fossil resources with microbial fermentation processes that utilize renewable biomass resources. Microorganisms have great potential to convert sugars to various useful compounds, such as alcohols, organic acids, and polymer materials, via multi-step enzymatic reactions, known as metabolic pathways. Because the role of native metabolism is not to produce our target compounds, its innate productivity is generally not so high. Therefore, modifying the metabolic system to enhance the productivity of the target compound is necessary. Many technologies have been developed to optimize metabolic pathways. In this lecture, I will present methodologies for the computational design of metabolic pathways using a stoichiometric model and for the experimental evaluation of metabolic flow in these pathways. In addition, methods for controlling metabolic flow on pathways to approach an ideal state will be introduced.

Bio: Yoshihiro Toya is an Associate Professor at The University of Osaka.
Assoc. Prof. Shunsuke Saruwatari
Shunsuke Saruwatari

Millimeter-level Wireless Localization Using UWB Wireless Networks

Time: Wed, Mar 11, 02:00 JST

Abstract: Ultra-Wideband (UWB) wireless technology, best known today for powering precise location tracking in consumer devices such as Apple's AirTag, enables radio signals to measure distances with centimeter-level precision. While current UWB systems require multiple anchors installed around a room, this setup is often impractical for everyday environments such as offices, homes, or factories. In this talk, I will introduce our research on millimeter-level localization using a single compact UWB module. Our system, developed as part of the XRLoc project, can track the position of multiple objects in real time from a single point, without the need for multiple anchors. We demonstrate how this technology can turn ordinary objects, such as cups or tools, into trackable interactive elements in extended-reality (XR) and robotics applications. By reducing infrastructure cost and setup complexity, this approach brings high-precision localization from specialized labs into everyday environments, bridging the gap between consumer devices like AirTags and industrial-grade positioning systems.

Bio: 2007 Ph.D. (Science), The University of Tokyo. 2008–2011 Assistant Professor, RCAST, The University of Tokyo, Japan. 2012–2015 Lecturer, Graduate School of Information Science, Shizuoka University, Japan. 2016–Present Associate Professor, Graduate School of Information Science and Technology, The University of Osaka, Japan. 2023–Present Visiting Scholar, ECE, the University of California, San Diego, USA. Specialization: Wireless Networks, Sensor Networks, IoT, System Software.
Assoc. Prof. Yuya Sasaki
Yuya Sasaki

Graph Data Analysis: From Efficiency and Accuracy to Trustworthy

Time: Wed, Mar 11, 17:00 JST

Abstract: Graph data are widely ubiquitous, ranging from social networks to molecules. Graph data analysis aims to efficiently and accurately analyze such data and to uncover new knowledge. The goal of this talk is to introduce core graph data analysis techniques: graph search, graph mining, and graph neural networks. Then, I will discuss how graph analytics can evolve beyond efficiency and accuracy toward trustworthy outcomes.

Bio: Yuya Sasaki is an Associate Professor at The University of Osaka.

Lab page
Prof. Taisuke Izumi
Taisuke Izumi

Distributed Algorithm Beyond Worst Case

Time: Thu, Mar 12, 17:00 JST

Abstract: The worst-case analysis, which evaluates the performance of algorithms with the most difficult inputs, is standard in theoretical computer science. However, real-world inputs are not always adversarial as assumed in worst-case analysis, and often have better properties arising from their application domain. This can cause a gap between the theoretical guarantees of algorithms and their practical performance. The Beyond-Worst-Case paradigm is a general term for algorithm design that aims to close this gap. By mathematically characterizing typical properties of real data and designing algorithms that exploit these properties, this paradigm explores theoretically rigorous algorithm design with practical efficiency. In recent years, the Beyond-Worst-Case paradigm has become an important research field in theoretical computer science, and has led to many new techniques and problem formulations. In this seminar, we present recent progress in Beyond-Worst-Case approaches in the field of distributed graph algorithms that studies the methodology of in-network solving of graph problems taking network topologies as inputs. Although the topics are highly theoretical and mathematical, no knowledge beyond undergraduate-level graph theory and algorithm theory is required.

Bio: 2006 Ph.D. (Computer science), Osaka University. 2006–2009 Assistant Professor, Nagoya Institute of Technology, Japan. 2009–2020 Associate Professor, Nagoya Institute of Technology, Japan. 2020–2025 Associate Professor, Osaka University, Japan. 2025–Present Professor, The University of Osaka, Japan. Speciality: Algorithm design and complexity analysis, distributed systems.
Assoc. Prof. Jun Shiomi
Jun Shiomi

Towards a Safe and Secure Society: Detection of Hardware Trojans in Integrated Circuits

Time: Thu, Mar 12 & Fri, Mar 13, 17:00 JST

Abstract: Our modern lives are entirely dependent on Integrated Circuits (ICs) found in everything from smartphones and cars to critical medical equipment. While these chips are the heart of modern society, they face a severe, invisible threat, particularly because they are manufactured across a global supply chain involving multiple companies. This threat is called the Hardware Trojan, a malicious circuit secretly inserted during the chip design or manufacturing process. It poses an existential risk, capable of causing devastating data leaks or catastrophic system failure, fundamentally undermining our privacy and the safety of critical social infrastructure. This talk will firstly provide a simple overview of how ICs are designed and how Hardware Trojans are injected. It will then introduce cutting-edge techniques to detect Hardware Trojans and novel Hardware Trojans for emerging applications. The goal of this talk is to provide the audience with a better understanding of security regarding their daily technology and highlight the efforts required to build a trustworthy information society.

Bio: Jun Shiomi is an Associate Professor at The University of Osaka.