Our laboratory studies basic software theory and its applications to programming languages, program verification, and program transformation. Recently, we have been especially focusing on automatic program verification methods based on higher-order model checking.

We conduct research on user interfaces to make computer applications easier to use. In particular, we are conducting the following research

1. Research on interaction methods that enable easy generation and use of 3D graphics

2. Research on pen input interface utilizing the freedom of handwriting

3. Research on interfaces for efficient handling of huge information spaces

Our laboratory conducts a variety of research on machine learning.

1. Basic theory of supervised learning, unsupervised learning, reinforcement learning, etc.

2. Practical algorithms for supervised learning, unsupervised learning, reinforcement learning, etc.

3. Applications of machine learning techniques to industry and natural sciences

Our research focuses on the following research themes in natural language processing and computational linguistics.

1. Computational models and basic analysis techniques for natural language: syntax and semantics, syntactic analysis, semantic analysis, etc.

2. Applications of natural language processing: machine translation, question answering, dialogue systems, etc.

3. Grounding to connect natural language and other media (images/video, numerical data, etc.)

We study the following topics in machine learning.

Theoretical analysis of learning, including quantifying prediction uncertainty, learning feature representations of data, learning based on human interaction, learning from biased learning data, and deep learning.

As an application, we are developing analysis and support systems for the medical field.

Our research focuses on the following topics in algorithms and discrete mathematics.

1. (Theoretical, Scalable) Graph Algorithms (Graph cuts, Graph coloring etc)

2. Generalizations of the Four Color Theorem

3. Applications of Graph algorithm/graph theory to Machine Learning (GNN etc)

We study system software in general, primarily operating systems and virtualization software. We also study computer security, either leveraging system software or targeting the system software itself.

Computational science is a new approach to science following experimental and theoretical approaches that has developed along with the exponential growth of computers. In this laboratory, we conduct education and research at the boundary between computational science and computer science, with first-principles electronic structure calculations at the core.

Focusing on computer architecture, we will promote research on highly efficient and enjoyable next-generation computers, including custom computing using FPGAs and dedicated hardware, co-design of algorithms to be computed and hardware such as machine learning processing, and high-level synthesis compilers to support hardware design.

Our laboratory conducts research in image processing and spatio-temporal data analysis. In particular, we are working on the following topics

1. Image processing: super-resolution, image restoration, denoising, compressed sensing image reconstruction, image synthesis

2. Spatio-temporal data analysis: semantic domain segmentation, change detection and recognition, signal source separation

3. Applications of image processing and spatio-temporal data analysis: disaster situational awareness, environmental monitoring

Our lab focuses on two main areas: (1) software theory and software engineering, where we aim to design novel methodologies, engineering techniques, and toolchains for building trustworthy intelligent software systems; and (2) adapting and adopting intelligent software techniques to diverse cyber-physical and cyber-cyber systems across various industrial domains to solve real-world challenges.

Based on the fusion of machine learning and symbolic logic, we are engaged in research on computational linguistics and natural language processing, focusing on the following topics, aiming at "natural language processing technology that understands the meaning of natural language more like humans".

1. Interdisciplinary and multidisciplinary analysis of statistical language processing techniques

2. Fundamentals and applied techniques of semantic analysis and natural language inference