Posters

Poster Abstracts

Amy BILTON - Overcoming the Barriers of Sustainable Technology Adoption in the Developing World Using Sensors, Data, Artificial Intelligence, and Human-Centered Design
Sacha RUZZANTE, Eren RUDY, Yu CHEN, and Amy M. BILTON

Many organizations are focused on the development of new technologies to achieve the United Nations Sustainable Development Goals (SDGs) and improve the quality-of-life of marginalized populations in the developing world.  Despite the wealth of new products developed and the efforts put in by development teams, many are not adopted by the target populations.  The Water and Energy Research Laboratory (WERL) at the University of Toronto is developing multi-disciplinary approaches to address this complex issue.  This poster overviews a study which examined variables that regularly explain adoption across a range of developing world agricultural technologies and contexts. We conducted a meta-analysis of 367 regression models from adoption studies in the published literature. We found that, on average, farmer education, household size, land size, access to credit, land tenure, access to extension services, and organization membership positively correlate with the adoption of many agricultural technologies. Based on these learnings we provide some recommendations for adoption researchers and policy makers, but, given the variability of the results, conclude that efforts to promote agricultural technologies in the developing world must be adapted to suit local contexts.  In line with this learning, this poster also overviews mixed-method approaches currently being undertaken to understand adoption barriers for water and sanitation systems.  These approaches leverage data which can be gathered on user-behaviour using unobtrusive embedded sensors to fill in gaps which exist with traditional social science methods.  By developing methods and tools to understand user motivations for product use, we can improve the system design to enable products to have more impact on those who need it most.
 

Yu FUKASAWA - Electrical conversation of wild mushrooms after the rain
Yu FUKASAWA 1* , Daisuke AKAI 2 , Masayuki USHIO 3,4,5 , Takayuki TAKEHI 2
1 Graduate School of Agricultural Science, Tohoku University, 232-3 Yomogida, Naruko, Osaki, Miyagi 989-6711, Japan
2 National Institute of Technology, Nagaoka College, 888 Nishi-Katakaimachi, Nagaoka, Niigata 940-0817, Japan
3 Hakubi Center, Kyoto University, Kyoto 606-8501, Japan
4 Center for Ecological Research, Kyoto University, Otsu 520-2113, Japan
5 Present address: Department of Ocean Science, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong SAR

We measured extracellular bioelectrical activities of wild mushrooms of Laccaria bicolor on forest floor to examine its response to some environmental factors and the potential for signal transport across the mycelial network. Six fruit bodies of L. bicolor in a cluster, to which electrodes were attached, exhibited less electrical potentials at the beginning, probably due to the lack of precipitation for over a week. However, its electrical potential fluctuated after raining, sometimes over 100 mV. The electrical potential of the fruit bodies and its fluctuation were correlated with precipitation and temperature. Causality analysis of electrical potential after the rain showed
electrical signal transport among fruit bodies, particularly between spatially close ones, with potential directionality. Our preliminary results bring a call for studies on fungal electrical potentials in a more ecological context under field conditions.

Tetsuya HAMA - Interstellar cold dust surface chemistry: a key to understanding molecular evolution in space
Tetsuya HAMA

Recent astronomical observations have been successful in revealing that interstellar clouds, the birthplaces of stars and planets, are “chemically rich” despite their low-temperature (around 10 K) and low-pressure (typically 104 H2 molecules cm−3) environments.  Interstellar clouds consist of gases and submicrometer-sized dust grains made from amorphous silicates or carbonaceous material. To date, around 280 molecules have been discovered in the gas phase. In addition, the surface of dust grains is covered by the solid forms of molecules such as H2O, CO2, NH3, CH4, H2CO, and CH3OH in interstellar clouds. These icy dust grains, which are often called “ice mantles”, have played a critical role in the formation of icy giants in the early solar system as a dominant component of their building blocks.

Astronomers want to know how and when interstellar molecules formed in interstellar clouds. In addition, many chemists are fascinated by physicochemical processes under extreme conditions, i.e., at the very low temperatures and ultrahigh vacuum associated with interstellar clouds. For understanding the formation of the solar system, laboratory study of the surface chemistry on interstellar dust is highly desirable.

In my poster, I present an overview of physicochemical surface processes on interstellar dust, with emphasis on our recent experiments investigating the quantum tunneling reaction of H atoms on the ice surface around 10 K.

References: T. Hama, and N. Watanabe, Chemical Reviews 113, 8783 (2013).  T. Hama, A. Kouchi, and N. Watanabe, Science 351, 65 (2016).

Yuki HIBIYA - Origin of the short-lived radionuclide 92Nb and Solar System formation
Yuki Hibiya, Tsuyoshi Iizuka, Hatsuki Enomoto, and Takehito Hayakawa

The short-lived radionuclide niobium-92 (92Nb) has been used to estimate the timing of supernova explosions and planetary differentiation, assuming that it was uniformly distributed in the early solar system. Here, we present the niobium–zirconium (Nb–Zr) dating of Northwest Africa 6704, a meteorite thought to form in the outer protosolar disk due to nucleosynthetic isotope similarities with carbonaceous chondrites. The result defines an initial 92Nb/93Nb ratio of (2.72 ± 0.25) × 10−5 at the NWA 6704 formation, 4562.76 ± 0.30 million years ago. This corresponds to a 92Nb/93Nb ratio of (2.96 ± 0.27) × 10−5 at the time of solar system formation, which is ∼80% higher than the values obtained from meteorites formed in the inner disk. Our newly obtained initial 92Nb/93Nb value is higher than the expected galactic background produced solely by thermonuclear Type Ia supernovae, requiring that 92Nb has a core-collapse supernovae (CCSNe) origin. Given that short-lived radionuclides inherited from the interstellar medium were homogeneously distributed in the protosolar disk, the 92Nb heterogeneity suggests that a nearby CCSN contributed significantly to 92Nb production. The enrichment of such CCSN ejecta in the outer disk could explain the enigmatic heterogeneity of 26Al and nucleosynthetic stable-isotope anomalies in the disk. We envisage that the CCSN ejecta was injected into the core of the protosolar cloud shortly before it collapsed, consistent with the hypothesis that a supernova triggered the formation of our solar system.

Norikazu ICHIHASHI - Toward the creation of life-like molecular systems through evolution and design
Norikazu ICHIHASHI, The University of Tokyo
 
Our research group aims to understand the principles that govern living systems by constructing life-like molecular systems. We have two approaches for construction, experimental evolution and design. In the experimental evolution, we performed a long-term Darwinian evolution of an artificial genomic RNA, which replicates itself through the translation of its encoded RNA replication enzyme in a test tube. We observed several life-like phenomena, such as the emergence of parasitic entities, diversification, and complexification, during the process. These observations informed us how a simple self-replication molecule develops complexity through evolution. In the design approach of life-like molecular systems, we are pursuing the realization of a self-regeneration system in a test tube. In this system, all biological macromolecules (DNA, RNA, and proteins) are designed to be produced using the same mechanism as in the cell. To date, we have achieved continuous DNA replication and partial regeneration of some translational factors. The realization of a self-regeneration molecular system has the potential to replace all processes that currently rely on natural living organisms. I would like to discuss the significance and potential impact of these studies on society with researchers from various fields.

Tomoko INOSE - Plasmon-based single live-cell nanowire endoscopy toward unraveling biological systems
 Masahiko YOSHIMURA, Ryuto SASAYAMA, Shuhei FURUKAWA, Hiroshi UJI-I, Tomoko INOSE
 
Physically addressing site-specific areas of a single live cell with the high spatial resolution is one of the attractive approaches to obtaining a better understanding of cellular behavior. Cell endoscopy is an attractive approach to realizing site-specific access inside a single live cell. In conventional cell endoscopy, several probes, such as glass pipettes, glass fibers, nanowires, and nanotubes, are inserted into a cell, allowing optical studies, material delivery, electrochemistry, and so on. Optical waveguide probes are increasingly popular because they enable optical information to be easily directed to, and extracted from, the cell.

This poster presentation will introduce plasmon-based single live-cell nanowire endoscopy as a novel tool for understanding biological processes at a single-cell level. Plasmonic properties on novel metal nanoparticles/nanowires can enhance Raman/fluorescence signals greatly. By applying this unique plasmonic property of a nanowire to a single live-cell endoscopic probe, we have realized highly sensitive sensing of Raman signals from site-specific areas in a single live cell with less invasiveness. This technique gives us essential information, such as the interaction between DNA and fluorescence dye inside a nucleus, or site-specific intracellular local pH. Toward further promising applications of this plasmon-based nanowire endoscopy, we have been challenged to apply this technique to intracellular materials delivery systems.

Setsu KATO - The addition of butanol causes the formation of aggregates and cell death in Escherichia coli
Setsu KATO, Misaki SHINZATO, Yuto ARAKI, Ryuji KAWABATA, Junya KATO, Yoshiteru AOI, Yutaka NAKASHIMADA

Butanol is a promising candidate for biofuels because of its favorable chemical characteristics. However, the fermentation of butanol is difficult because it has a strong cell toxicity. Thus, unraveling the toxic mechanism of butanol is essential in order to develop strategies to strengthen the weak points of cells under the stress. In this study, we investigated the cellular response of Escherichia coli at the single-cell level following the addition of 1-butanol to the cell culture. It turned out that the addition of butanol triggered the formation of aggregates in cells. Cells with aggregates were unable to proliferate in the fresh medium after butanol removal, indicating these cells were dead. According to the previous reports, the addition of butanol disturbs the cellular membrane homeostasis and, in some cases, results in the production of blebs. Therefore, we wondered if the formation of aggregates we observed was just a result of secondary effects triggered by the cell membrane disruption. To answer this question, we stained the butanol-treated cells with propidium iodide (PI) and compared the proportion of cells stained with PI to the proportion of aggregate-forming cells. The results showed that cells with aggregates existed more frequently than cells stained with PI, suggesting that the formation of aggregates itself could be the direct cause of cell death. This study revealed that butanol toxicity has at least two separate mechanisms to cause cell death.

Hikaru KURAMOCHI - Tracking chemical reaction dynamics by ultrafast nonlinear spectroscopy using few-cycle pulses.
 Hikaru KURAMOCHI
 
One of the frontiers in modern chemical science today is to elucidate how functional complex molecules, such as proteins, achieve sophisticated functionalities. Understanding their underlying molecular mechanisms provides not only a deeper understanding of the exquisite directionality and efficiency of particular chemical reactions they achieve but also design strategies for creating further sophisticated artificial molecular systems and materials. In this quest, it is desired to elucidate molecular dynamics from the reactant down to the product with electronic/structural details and with temporal resolution as high as possible. We develop advanced ultrafast laser spectroscopic techniques based on state-of-the-art optical technology and study chemical reaction dynamics in functional complex molecules. In particular, we use ultrafast nonlinear spectroscopy based on few-cycle pulses and track the change of the electronic/vibrational structure during the chemical reaction with a high temporal resolution down to 10 fs, aiming to find out necessary and sufficient molecular events for functioning. We also aim to realize such extreme ultrafast spectroscopic experiments at the single-molecule level in the condensed phase at room temperature. In the presentation, I will overview our recent studies on ultrafast structural dynamics in photochemistry and some ongoing attempts.

David C. LEITCH - Chemical Cartography: Mapping Chemical Structure and Reaction Space using Multivariate High-Throughput Experimentation
David C. LEITCH
 
The landscape of chemical structure and reaction space is astronomically vast, with countless possible chemical structures and synthesis routes to prepare them. This is a prevailing challenge specifically in the field of drug discovery/development, where the difficulty of identifying suitable chemical matter is intimately linked to the difficulty of efficiently preparing candidate compounds. To address this challenge, many researchers are applying advanced data analysis techniques and automated experimentation to better understand and predict outcomes in this space. The Leitch lab approach uses a combination of physical organic chemistry, high-throughput experimentation, and multivariate statistical analysis to tackle problems in catalysis, predictive modeling, and Active Pharmaceutical Ingredient synthesis. We are able to generate our own high-quality experimental data sets targeted to specific research questions, and use these to identify new chemical reactions, access new chemical structures, and predict structure-reactivity relationships in key organic chemistry transformations. Ultimately, we aim to streamline the process of synthesis planning predictions for complex molecule targets, and provide new tools and methods to efficiently access these targets.

Xiaoxiao LI - Learning without Forgetting?! Class Impression for Data-free Incremental Learning 
Sana AYROMLOU, Purang ABOLMAESUMI, Teresa TSANG, and Xiaoxiao LI

'Standard' deep learning-based classification approaches require collecting all samples from all classes in advance and are trained offline. This paradigm may not be practical in real-world clinical applications, where new classes are incrementally introduced through the addition of new data.  Class incremental learning is a strategy allowing learning from such data. However, a major challenge is catastrophic forgetting, i.e., performance degradation on previous classes when adapting a trained model to new data. Prior methodologies to alleviate this challenge save a portion of training data and require perpetual storage of such data that may introduce privacy issues. Here, we propose a novel data-free class incremental learning framework that first synthesizes data from the model trained on previous classes to generate a `Class Impression'. Subsequently, it updates the model by combining the synthesized data with new class data. To do so, we incorporate a cosine normalized Cross-entropy loss to mitigate the adverse effects of the imbalance, a margin loss to increase separation among previous classes and new ones, and an intra-domain contrastive loss to generalize the model trained on the synthesized data to real data. We compare our proposed framework with state-of-the-art methods in class incremental learning, where we demonstrate improvement in accuracy for the classification of 11,062 echocardiography cine series of patients. 

Takashi MATSUBARA - Geometric and Bayesian Deep Learning for Incorporating Our Needs
Takashi MATSUBARA (Osaka University, Japan)

Recent advancements in artificial intelligence technology are closely tied to the progress of deep learning, an inductive approach that approximates a system of interest using data. While it excels at creative areas, such as drawing, it struggles in areas that require rigor, such as natural and social sciences. This is mainly due to the scarcity and bias of available data, resulting in unrealistic, unreliable, and unfair decision making.

In contrast, scientific computing takes a more deductive approach. Its purpose is to make accurate predictions of real-world phenomena on a large scale and over a long period of time. To achieve this, it is necessary to gain insights into the mathematical structures underlying the phenomena and design computational methods that preserve those structures. This perspective defines convolutional neural networks as a computational method designed to preserve translational symmetry.

Related researchers including me propose to incorporate the idea of structure preservation into deep learning, which is called "geometric deep learning." This involves constraining the function space approximated by deep learning to a subspace. By doing so, when learning from data of physical phenomena and conducting simulations, it is possible to guarantee that energy conservation and momentum conservation laws are preserved. Moreover, this approach can help determine whether the phenomenon is energy-conservative or dissipative. In a more general sense, when applying medical data, disease-specific features can be selectively separated, which enables fair decisions that are not affected by individual differences such as gender and age.

Takahiro NISHIMICHI - Simulation-based large-scale structure cosmology: Emulation and beyond
Takahiro NISHIMICHI

Cosmological large-scale structures are shaped by nonlinear processes mainly driven by gravity. Previous methods of analyzing this rely on theoretical templates that are based on perturbative expansion about the linear solution, restricting the extraction of information to large, mildly nonlinear, scales. On the other hand, N-body simulations can uncover structures on smaller scales, until non-gravitational effects
such as gas cooling and feedback eventually become a factor.  However, their high computational cost hinders their direct use in statistical inference.

In this poster, I discuss the emulator approach as a potential solution. In particular, I present our Dark Quest simulation project and its applications to the SDSS and HSC datasets. A brief discussion on the future direction of the whole analysis framework is also presented, which involves  coupling between simulators and observations for automated knowledge acquisition.

Mikako OGAWA - Cancer targeted phototherapy, based on photo-chemical reaction
Mikako OGAWA, Hokkaido University

Photoimmuno therapy (PIT) is a new molecular-targeted phototherapy in which an antibody (Ab) conjugated with IR700, a hydrophilic silicon phthalocyanine derivative, is administered followed by irradiation with near-infrared light. When antibody-IR700 conjugates are bound to their target cells and are exposed to NIR-light, target cells rapidly undergo necrotic/immunogenic cell death in a highly selective manner. Real time microscopy demonstrates swelling, blebbing and bursting of the target cell membrane within minutes of light exposure with minimal damage to adjacent non-target cells. When exposing NIR light, physical stress was thought to be induced within the cellular membrane leading to increases in transmembrane water flow that eventually lead to cell bursting and immunogenic cell death (ICD).
One of the characteristics of PIT is that the antibody-IR700 complex only needs to bind to the surface of cancer cells, and the drug does not need to be internalized into the cells.  The formation of aggregates of IR700 on the cell membrane by photochemical reaction is an important mechanism of cell killing. That is, water-soluble axial ligands of IR700 are cleaved by the photochemical reaction, and the phthalocyanine stacks up due to the π-π interaction, resulting in the formation of aggregates. In addition, it was recently found that the formation of radical anions of IR700 and their protonation are essential for the progress of this photochemical reaction. The elucidation of these mechanisms may lead to the development of more effective compounds.
 

Hiroko OKUDAIRA - Parental Investment After Adverse Event: Evidence from the Great East Japan Earthquake
Tomohiko INUI, Hiroko OKUDAIRA

Parents increase private investment in their children when they fear the negative effects of an adverse event. Such an endogenous response makes it difficult to identify the cost of the adverse event and those disadvantaged by the shock. This study investigates the nature of an adverse shock that leads to endogenous responses by parents. Relying on the types of damage caused by the Great East Japan Earthquake, we find that the parents exposed to intense ground motion increased investment in their children’s cognitive skills. This positive response was not driven by attrition, migration, local demand, and supply shocks, or other mediating factors such as declines in home education or the extent of radioactive concentration. We also find that the exposure to intense ground motion increased parents’ aspirations for their children’s higher education. We interpret that the traumatic experience shifted the parents’ preferences toward investment in the children’s human capital. 

Kazumi OZAKI - Coupled evolution of the atmosphere and life before the advent of oxygenic photosynthesis
Kazumi OZAKI

The activity level of the biosphere is a crucial factor controlling the chemical composition of the atmosphere, but the quantitative understanding of how primitive anaerobic life has affected Earth’s habitability is still lacking,  particularly in the early Archean before the emergence of oxygenic photosynthesis. Here, we reconstruct global biological activity and its impacts on atmospheric chemistry by employing a numerical model of global biogeochemical cycles of H, C, and Fe on the early Earth. We explored the impact of varying key model parameters on global biogeochemistry using a stochastic approach and searched for possible solutions which allow for a habitable climate state. The results suggest that net primary production (NPP) was ~0.1% and ~1% that of the modern Earth for the non-photosynthetic (chemotrophic) ecosystem and anoxygenic photosynthetic ecosystem, respectively. The extremely low NPP fluxes produced by our model imply that geological fluxes of reductants, rather than nutrients (e.g., phosphorus), would have been the limiting factor for biological productivity. Despite the limited productivity of the primitive biosphere, atmospheric methane concentrations would have been relatively high (~300 ppmv) because primitive biospheres efficiently recycled the material in the ocean-atmosphere system. Our results also suggest that the transition from a chemotrophic ecosystem to an anoxygenic photosynthetic ecosystem exerts minor impacts on atmospheric composition. A better mechanistic understanding of the coupled evolution of primitive life and the atmosphere has great ramifications, not only for Earth's sustained habitability, but also for the search for life on Earth-like exoplanets with reducing atmospheres that may harbor primitive life forms.

Yuki SAKURAI - Exploring the Origin of the Universe ~ Cosmic Microwave Background Experiment ~
Yuki SAKURAI (Okayama University)

The origin of the universe is one of the greatest mysteries for human beings. An important cosmological theory, “cosmic inflation” is deeply related to this mystery, which predicts a rapid exponential expansion in the early universe. The theory can explain several unsolved problems in cosmology, making it an important and promising candidate. However, it has not been verified experimentally.

The inflation theory predicts that primordial gravitational waves are generated during inflation due to a space-time distortion. Through  interference with primordial gravitational waves, characteristic patterns are imprinted in the polarization of the cosmic microwave background (CMB) radiation. It is conclusive evidence of inflation and thus allows for experimental verification by precise observation of CMB polarization.

 The CMB is the oldest electromagnetic wave in the universe that is still isotropically observable in the full sky even today. Therefore, CMB experiments aiming to verify inflation are actively proceeding and planned around the world, on the ground, in balloons, and by satellite. Since the polarization that is evidence of inflation is a tiny signal (~nano Kelvin order), the observation requires the world's most advanced technologies in terms of both hardware and software. LiteBIRD is an international satellite project to search inflation based on precise full-sky CMB polarization observations, aiming to launch around 2030. The project is led by Japan with the participation of Canada. In this presentation, we introduce the science, the current status, and the instrument of CMB experiments using LiteBIRD as a representative example.

Taikan SUEHARA - High-Level Event Reconstruction with Graph Neural Network for Future Colliders
Taikan SUEHARA, Shusaku TSUMURA, Tomoki ONOE, Yuta NAKASHIMA, Noriko TAKEMURA, Hajime NAGAHARA

A collider detector is a gigantic assembly of particle detectors such as silicon sensors, gas tracking devices and calorimeters. Modern detector designs feature a huge number of channels of finely-segmented sensor components and integrated electronics to dissolve precise characteristics of hundreds of particles emerging at the interaction point of colliders. Reconstructing those characteristics precisely from millions of sensor channels is a challenging task and it thus can be a good application of modern machine learning. Future colliders, such as the International Linear Collider (ILC) project considered to be built in Japan with international cooperation, and their detector designs are well suited to pursue the possibility of using cutting-edge technologies of machine learning.

Graph Neural Network (GNN) is considered to be a good structure for such tasks since the input is basically a collection of sparsely-distributed 3D points with some features (such as energy deposit and arrival time). We will introduce two ongoing studies of the ILC related to GNNs. First is the application to clustering of calorimeter hits for reconstruction of jets. The algorithm, GravNet, was developed for the reconstruction of an upgrade of an ongoing collider project (HL-LHC) and was applied to ILC calorimeter simulation. Another is the application to quark flavor tagging, which aims to replace current algorithms by combining secondary vertex finding and flavor tagging with Graph Attention (GAT) network with three types of outputs. The status and future plans of those works will be presented.

Mitsuaki TAKEMI - Toward Trusted Neurotechnology – Development of Neurotech Guidebook and Evidence Book
Mitsuaki TAKEMI, Graduate School of Science and Technology, Keio University, Japan.
 

In recent years, neurotechnology, or so-called “neurotech,” which is the technology that aims to estimate and regulate the state of the human brain, is spreading rapidly. Brochures for neurotech products for general consumers make enticing claims such as “improved attention and focus,” “improved sleep quality,” or “improved athletic performance.” However, is there any scientific basis for these claims? Are there any dangers in ordinary consumers using such products on their own? Neurotech is a developing technology whose efficacy and safety have not yet been fully clarified. Given this situation, we published the Braintech guidebook in Japanese (https://doi.org/10.14991/KO52004001) last year to accurately share the current status and challenges of neurotech with the general public. Apart from this, we are currently developing an “evidence book” that summarizes the scientific evidence for the common claims made by neurotech products and the safety of their use. To this end, an Evidence Evaluation Committee of more than 40 researchers specializing in neuroscience is conducting systematic reviews on neurofeedback, neuromodulation, and EEG biomarkers. So far, four systematic reviews have been completed, including a preprint of a review on neurofeedback for motor performance improvements in healthy adults (Onagawa et al., bioRxiv 2022). The results of Onagawa et al. (2022) indicate no conclusive evidence to support the beneficial effects of neurofeedback on motor performance due to publication biases and substantial heterogeneity across studies. More empirical neurofeedback studies are needed to demonstrate its beneficial effects on motor performance and to safely incorporate neurofeedback into real-world scenarios.

Daisuke TANAKA - Machine-Learning-Assisted Exploration of Metal-Organic Frameworks
Daisuke TANAKA

Machine learning has recently attracted attention as an efficient exploration tool, especially in the area of materials science. In particular, several attempts to optimize synthesis conditions of materials using machine learning have been reported. However, the application of machine learning techniques to predict synthesis conditions for novel materials remains limited while its application in novel material discovery seems promising. One reason for this is that machine learning is generally good at interpolation, but the search for synthesis conditions for new materials essentially involves extrapolation because it is necessary to try completely new synthesis conditions that are different from those used in the past. Another reason is that it is difficult to quantify the synthesis of new materials, making it difficult to treat the problem of synthesis condition search as a regression problem.

In this presentation, I will talk about synthetic exploration of novel metal-organic frameworks (MOFs) assisted by machine learning. MOFs are porous organic-inorganic hybrid materials, which have been widely studied as promising environmental energy materials. To the best of our knowledge, this is the first example of applying machine learning to synthesis for the exploration of unknown MOFs. The synthetic results generated from high-throughput screening experiments were evaluated using interpretable machine-learning techniques, which enabled us to determine the optimal conditions for the synthesis of novel MOFs.

Yu YOKOI - Algorithmic Study of Matching under Preferences
Yu YOKOI

The theory of matching under preferences is a mathematical framework to investigate matching problems involving agents with preferences. The field lies at the intersection of computer science and economics (game theory and mechanism design) and aims at designing algorithms that calculate fair and efficient matchings. Because matching problems are combinatorial in nature, naive brute-force type methods take an unreasonable amount of execution time (known as "combinatorial explosion"). Therefore, it is necessary to analyze the mathematical structures of the problems and design algorithms that exploit those structures. I have designed algorithms for constrained matching problems using tools from discrete mathematics, such as matroid theory and discrete convex analysis. In this poster, I give an overview of the field and present a part of my research.