Kazunari Wada

ORCID: 0009-0005-8260-4493
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Model Reduction and Neural Networks
  • Computer Graphics and Visualization Techniques
  • Topology Optimization in Engineering
  • Optical measurement and interference techniques
  • Thin-Film Transistor Technologies
  • Silicon and Solar Cell Technologies
  • Advanced Vision and Imaging
  • 3D Shape Modeling and Analysis
  • Heat Transfer Mechanisms
  • GaN-based semiconductor devices and materials
  • Electrostatic Discharge in Electronics
  • Acoustic Wave Resonator Technologies
  • Semiconductor materials and interfaces
  • Advanced Multi-Objective Optimization Algorithms
  • Semiconductor Quantum Structures and Devices
  • Generative Adversarial Networks and Image Synthesis
  • Strong Light-Matter Interactions
  • Quantum and electron transport phenomena

The University of Tokyo
1994-2024

Tohoku University
1999

NTT Basic Research Laboratories
1994

Development Bank of Japan
1994

The radiative decay of one-dimensional (1D) excitons is studied in novel GaAs quantum wires (QWI's) prepared by molecular beam epitaxy. temperature dependence the lifetime \ensuremath{\tau}(T) measured, and compared with that 2D wells (QW's): 1D grows much more slowly T than excitons, resulting shorter for T>20 K. This ascribed to thermalization concentrated density states QWIs. Analysis based on a theory has revealed expressed as 92\ensuremath{\surd}T ps...

10.1103/physrevlett.72.924 article EN Physical Review Letters 1994-02-07

10.1016/j.cma.2024.116746 article EN Computer Methods in Applied Mechanics and Engineering 2024-01-31

Generative adversarial networks (GAN) have recently been used for a design synthesis of mechanical shapes. A GAN sometimes outputs physically unreasonable For example, when model is trained to output airfoil shapes that indicate required aerodynamic performance, significant errors occur in the performance values. This because only considers data but does not consider equations lie under data. paper proposes physics-guided training guide learn physical validity. Physical validity computed...

10.48550/arxiv.2308.10038 preprint EN other-oa arXiv (Cornell University) 2023-01-01

Multiple shapes must be obtained in the mechanical design process to satisfy required specifications. The inverse problem has been analyzed previous studies obtain such shapes. However, finding multiple a short computation period is difficult while using conventional methods. This paper proposes use of conditional variational autoencoders (CVAE) with normal distribution, denoted by N-CVAE, along von Mises-Fischer S-CVAE, find solutions for problems. Both CVAE models embed into latent space....

10.48550/arxiv.2106.09901 preprint EN cc-by arXiv (Cornell University) 2021-01-01

Crystallinity of area-selective Ge layer with a (0 0 1) surface grown on Si substrate has been investigated by means diffractometry using parallel X-ray microbeam. The measured lattice parameters 〈0 1〉 direction were about 0.17% smaller than that bulk crystal. This tensile strain value was almost the same as simulated ones used finite-element method.

10.1017/s2044820110000559 article EN Diamond Light Source Proceedings 2010-11-01

10.7567/ssdm.2012.a-7-5 article EN Extended Abstracts of the 2020 International Conference on Solid State Devices and Materials 2012-09-27

We have developed aluminum nitride (AlN) epitaxial growth technology using Knudsen pressure MOCVD method. The thickness uniformity was ±1%. However groove-like cracks were formed on the surface of AlN film. deposition off-angle substrates and at high temperature been investigated for eliminating surface. an Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> which is -4 degree from...

10.1109/ultsym.1999.849399 article EN 1999-01-01

There has been research on airfoil design using deep generative models such as adversarial network (GAN). However, in prior methods, the generated results do not always satisfy governing equations. This paper reports of an attempt to construct a physics guided model and use it for fine-tuning. Computational software that calculates aerodynamic performance shapes was placed network. An objective function expressed terms relationship between required performance. As result, is confirmed...

10.1299/jsmeoptis.2022.14.u00057 article EN The Proceedings of OPTIS 2022-01-01
Coming Soon ...