FY 2023 RIEC Annual Meeting on Cooperative Research Projects|RIEC, Tohoku University|Center for Nation-Wide Cooperative Research on ICT

Program Outline

Speakers

Opening Session 10:00-10:15

Date 10:00-10:15

RIEC and its cooperative research projects

HANYU Takahiro Director / Professor

RIEC, Tohoku University

Session 2 Computing systems with transcendent computational capability 10:15-11:15

Date 10:15-10:45

Integrated synthesis of graphene nanoribbon-based quantum dot devices

KATO Toshiaki Associate Professor

Graduate School of Engineering, Tohoku University

Recently, graphene nanoribbon-based quantum dot (GNR-QD) attracts intense attention due to its excellent electrical characteristics.
In this study, we developed unique method, which can selectively and directly grow GNR-QD at the specific position of pre-formed electrical devices. It is also proved that our fabricated GNR-QD device shows outstanding QD features.
Since integrated synthesis of GNR-QD devices becomes possible with our method, it can be expected to accelerate developing future high-performance quantum-information devices with GNR.

[Keywords] Graphene nanoribbon, Quantum dot, Integration

Date 10:45-11:15

Extended free energy model: super-hierarchal linkage of device function and structure

KOTSUGI Masato Professor

Advanced Engineering, Tokyo University of Science

We discuss automating magnetic domain structure interpretation using an 'Extended Landau Free Energy Model' in machine learning. It addresses the challenge of analyzing magnetic functionalities in nanoscale ferromagnets with complex interactions. The new model reveals the origins of energy barriers impeding information recording and visualizes their spatial concentration. This physically grounded approach is anticipated to elucidate mechanisms in various materials.

[Keywords] Data science, Free energy, Magnetics

Session 3 Information infrastructures as natural as breathing11:15-12:15

Date 11:15-11:45

A study on the rectenna array for power reception and propagation of energy for IoTs

MARUYAMA Tamami Professor

Department of Production Systems Engineering, National Institute of Technology, Hakodate College

The presenter has previously noted that when arranging a rectenna array using half-wavelength dipole antennas, each element functions as a rectenna, receiving power while simultaneously serving as a directors in a Yagi-Uda antenna to transmit power to adjacent elements. This phenomenon possesses an intriguing characteristic: the greater the number of power-receiving elements, the greater the distance over which power can be transmitted. In this presentation, we share simulation and experimental results demonstrating the application of this phenomenon to power the IoT terminals. The goal is to minimize size and enhance flexibility in terms of shape and layout.

[Keywords] Rectenna, Energy harvesting, IoT

Date 11:45-12:15

Application to advanced optical sensor based on SOI photodiode with SP antenna

SATOH Hiroaki Associate Professor

Research Institute of Electronics, Shizuoka University

SOI (silicon-on-insulator) photodiode with SP (surface plasmon) antenna,which is composed of silicon thin film and periodic metal diffraction grating, has high sensitivity for a specific incident condition, if the diffracted light could be resonantly coupled with the SOI waveguiding mode. As a result, it allows us to fabricate the unique photodetectors with the light selectivities for desired wavelength, incident angle, and polarization by appropriately designing SP antenna and photodiode body. In this presentation, the fundamentals of SOI photodiode with SP antenna are explained, and two applications to angle-sensitive pixel (ASP) and biosensor as advanced optical sensors are also introduced.

[Keywords] Silicon photodiode, Angle-sensitive pixel, Biosensor

Session 4 Super-intelligent systems based on human understanding13:30-14:30

Date 13:30-14:00

Construction of the cortical representation of natural objects --- from pixels to objects

SAKAI Ko Professor

Department of Computer Science, University of Tsukuba

Along the cortical visual pathway, pixel-wise images are gradually transformed into objects and categories. Such hierarchical transformation has also been widely known in Deep Convolutional Neural Networks. The mid-level visual area, V4, has been considered to contribute to the transformation, and construct the intermediate-level representation of objects. To clarify the encoding in V4 regarding object shapes and surface properties, we analyzed the spike recordings from Monkey V4 while the animals were viewing a set of natural images. The results indicated that individual neurons simultaneously code multiple features and their population selectively bind the features to construct object-based representation.

[Keywords] Vision, Object representation, Electrophysiology

Date 14:00-14:30

Verbal and nonverbal communication in human-robot interaction

HUANG Tsung-Ren Associate Professor

Department of Psychology, National Taiwan University

Social robots are becoming part of our everyday life. However, they are not yet as emotionally sensitive or socially intelligent as we like. In this lecture, I will discuss our recent studies that attempt to leverage verbal signals (e.g., linguistic and paralinguistic information) and nonverbal signals (e.g., facial expressions or other body languages) to improve a robot's understanding of humans, namely robot "Theory of Mind." Such an understanding can help personalization and maintain a long-term, human-robot relationship. Also, I will discuss communication issues in repeated human-robot interaction, such as unavoidable social contagion of false memories from robots to humans. Overall, the results from these studies can inform us how to make human-centered social robots, which respect not only individual differences but also limited cognitive capacity in humans.

[Keywords] Human-robot interaction, Communication, Human-centered design

RIEC Award Ceremony14:40-15:10

FY2023 RIEC Awardees were announced

FY2023 RIEC Awardees (Japanese site only)

Poster Session15:30-17:30

Poster Session Venue : RIEC Main Building 1F

Poster Session List

Reception17:40-18:40

Reception Venue : RIEC Main Building 1F, Open Seminar Room
Reception Fee Required(cash only) : 3,000 yen for non-students, 1,000 yen for students (Please pay on the day)

Registration by Google form Registration by E-mail

Research Institute of Electrical Communication, Tohoku University
Center for Nation-Wide Cooperative Research on ICT