-
Virtual Human Generative Model: Masked Modeling Approach for Learning Human Characteristics
Authors:
Kenta Oono,
Nontawat Charoenphakdee,
Kotatsu Bito,
Zhengyan Gao,
Yoshiaki Ota,
Shoichiro Yamaguchi,
Yohei Sugawara,
Shin-ichi Maeda,
Kunihiko Miyoshi,
Yuki Saito,
Koki Tsuda,
Hiroshi Maruyama,
Kohei Hayashi
Abstract:
Identifying the relationship between healthcare attributes, lifestyles, and personality is vital for understanding and improving physical and mental conditions. Machine learning approaches are promising for modeling their relationships and offering actionable suggestions. In this paper, we propose Virtual Human Generative Model (VHGM), a machine learning model for estimating attributes about healt…
▽ More
Identifying the relationship between healthcare attributes, lifestyles, and personality is vital for understanding and improving physical and mental conditions. Machine learning approaches are promising for modeling their relationships and offering actionable suggestions. In this paper, we propose Virtual Human Generative Model (VHGM), a machine learning model for estimating attributes about healthcare, lifestyles, and personalities. VHGM is a deep generative model trained with masked modeling to learn the joint distribution of attributes conditioned on known ones. Using heterogeneous tabular datasets, VHGM learns more than 1,800 attributes efficiently. We numerically evaluate the performance of VHGM and its training techniques. As a proof-of-concept of VHGM, we present several applications demonstrating user scenarios, such as virtual measurements of healthcare attributes and hypothesis verifications of lifestyles.
△ Less
Submitted 14 August, 2023; v1 submitted 18 June, 2023;
originally announced June 2023.
-
Disappearance and Survival of Superconductivity in FeSe under High Pressure
Authors:
Kiyotaka Miyoshi,
Shota Yamamoto,
Atsushi Shiota,
Takuya Matsuoka,
Masaki Ohe,
Yumi Yamamoto,
Shijo Nishigori
Abstract:
Superconductivity in FeSe was investigated under high pressure through the measurements of DC magnetization by using a diamond anvil cell. We successfully observed that the disappearance of the superconductivity as a results of the appearance of the non-superconducting ortho I\hspace{-1pt}I phase above $\sim$7 GPa ($\sim$5 GPa), when Ar (glycerin) is used as the pressure transmitting media. Contra…
▽ More
Superconductivity in FeSe was investigated under high pressure through the measurements of DC magnetization by using a diamond anvil cell. We successfully observed that the disappearance of the superconductivity as a results of the appearance of the non-superconducting ortho I\hspace{-1pt}I phase above $\sim$7 GPa ($\sim$5 GPa), when Ar (glycerin) is used as the pressure transmitting media. Contrary to this, it has been found that the superconductivity even survives under pressure above 7 GPa, when the thickness ($t$) of a platelet-single crystal specimen is reduced. The survival of the superconductivity above 7 GPa is consistent with a previous observation under hydrostatic pressure by using a cubic anvil apparatus, suggesting that the hydrostaticity of the pressure is improved by reducing $t$. It is also inferred that the appearance of the ortho I\hspace{-1pt}I phase is due to uniaxial stress along the [001] direction.
△ Less
Submitted 24 June, 2021; v1 submitted 30 March, 2021;
originally announced March 2021.
-
Global Data Science Project for COVID-19
Authors:
Toyotaro Suzumura,
Dario Garcia-Gasulla,
Sergio Alvarez Napagao,
Irene Li,
Hiroshi Maruyama,
Hiroki Kanezashi,
Raquel P'erez-Arnal,
Kunihiko Miyoshi,
Euma Ishii,
Keita Suzuki,
Sayaka Shiba,
Mariko Kurokawa,
Yuta Kanzawa,
Naomi Nakagawa,
Masatoshi Hanai,
Yixin Li,
Tianxiao Li
Abstract:
This paper aims at providing the summary of the Global Data Science Project (GDSC) for COVID-19. as on May 31 2020. COVID-19 has largely impacted on our societies through both direct and indirect effects transmitted by the policy measures to counter the spread of viruses. We quantitatively analysed the multifaceted impacts of the COVID-19 pandemic on our societies including people's mobility, heal…
▽ More
This paper aims at providing the summary of the Global Data Science Project (GDSC) for COVID-19. as on May 31 2020. COVID-19 has largely impacted on our societies through both direct and indirect effects transmitted by the policy measures to counter the spread of viruses. We quantitatively analysed the multifaceted impacts of the COVID-19 pandemic on our societies including people's mobility, health, and social behaviour changes. People's mobility has changed significantly due to the implementation of travel restriction and quarantine measurements. Indeed, the physical distance has widened at international (cross-border), national and regional level. At international level, due to the travel restrictions, the number of international flights has plunged overall at around 88 percent during March. In particular, the number of flights connecting Europe dropped drastically in mid of March after the United States announced travel restrictions to Europe and the EU and participating countries agreed to close borders, at 84 percent decline compared to March 10th. Similarly, we examined the impacts of quarantine measures in the major city: Tokyo (Japan), New York City (the United States), and Barcelona (Spain). Within all three cities, we found the significant decline in traffic volume. We also identified the increased concern for mental health through the analysis of posts on social networking services such as Twitter and Instagram. Notably, in the beginning of April 2020, the number of post with #depression on Instagram doubled, which might reflect the rise in mental health awareness among Instagram users. Besides, we identified the changes in a wide range of people's social behaviors, as well as economic impacts through the analysis of Instagram data and primary survey data.
△ Less
Submitted 3 August, 2021; v1 submitted 9 June, 2020;
originally announced June 2020.
-
Magnetoelectric Effect in the Antiferromagnetic Ordered State of Ce$_{3}$TiBi$_{5}$ with Ce Zig-Zag Chains
Authors:
Masahiro Shinozaki,
Gaku Motoyama,
Masahiro Tsubouchi,
Masumi Sezaki,
Jun Gouchi,
Shijo Nishigori,
Tetsuya Mutou,
Akira Yamaguchi,
Kenji Fujiwara,
Kiyotaka Miyoshi,
Yoshiya Uwatoko
Abstract:
In this study, we investigate the magnetoelectric (ME) effect of the newly discovered antiferromagnetic (AFM) compound Ce$_{3}$TiBi$_{5}$ with a hexagonal structure (space group $P6_{3}$/$mcm$). In this system, Ce ions form zig-zag chains that extend along the $c$-axis. We focus on the lack of the local inversion symmetry at the Ce site, although the crystal structure has an inversion center. A th…
▽ More
In this study, we investigate the magnetoelectric (ME) effect of the newly discovered antiferromagnetic (AFM) compound Ce$_{3}$TiBi$_{5}$ with a hexagonal structure (space group $P6_{3}$/$mcm$). In this system, Ce ions form zig-zag chains that extend along the $c$-axis. We focus on the lack of the local inversion symmetry at the Ce site, although the crystal structure has an inversion center. A theoretical study has predicted that magnetization can be induced by an electric current in the AFM ordering on the zig-zag chain. We conducted magnetization measurements on Ce$_{3}$TiBi$_{5}$ under an applied constant electric current and a static magnetic field around the AFM ordering temperature of $T_{\rm N}$ = 5.0 K. We successfully observed of the current-induced magnetization below $T_{\rm N}$. The magnitude of the current-induced magnetization has a linear electric current dependence and exhibits no magnetic field dependence. This behavior is consistent with the theoretical prediction of the ME effect in the ferrotoroidal ordered state.
△ Less
Submitted 27 February, 2020;
originally announced February 2020.
-
Macro Action Reinforcement Learning with Sequence Disentanglement using Variational Autoencoder
Authors:
Heecheol Kim,
Masanori Yamada,
Kosuke Miyoshi,
Hiroshi Yamakawa
Abstract:
One problem in the application of reinforcement learning to real-world problems is the curse of dimensionality on the action space. Macro actions, a sequence of primitive actions, have been studied to diminish the dimensionality of the action space with regard to the time axis. However, previous studies relied on humans defining macro actions or assumed macro actions as repetitions of the same pri…
▽ More
One problem in the application of reinforcement learning to real-world problems is the curse of dimensionality on the action space. Macro actions, a sequence of primitive actions, have been studied to diminish the dimensionality of the action space with regard to the time axis. However, previous studies relied on humans defining macro actions or assumed macro actions as repetitions of the same primitive actions. We present Factorized Macro Action Reinforcement Learning (FaMARL) which autonomously learns disentangled factor representation of a sequence of actions to generate macro actions that can be directly applied to general reinforcement learning algorithms. FaMARL exhibits higher scores than other reinforcement learning algorithms on environments that require an extensive amount of search.
△ Less
Submitted 3 June, 2019; v1 submitted 22 March, 2019;
originally announced March 2019.
-
FAVAE: Sequence Disentanglement using Information Bottleneck Principle
Authors:
Masanori Yamada,
Heecheol Kim,
Kosuke Miyoshi,
Hiroshi Yamakawa
Abstract:
We propose the factorized action variational autoencoder (FAVAE), a state-of-the-art generative model for learning disentangled and interpretable representations from sequential data via the information bottleneck without supervision. The purpose of disentangled representation learning is to obtain interpretable and transferable representations from data. We focused on the disentangled representat…
▽ More
We propose the factorized action variational autoencoder (FAVAE), a state-of-the-art generative model for learning disentangled and interpretable representations from sequential data via the information bottleneck without supervision. The purpose of disentangled representation learning is to obtain interpretable and transferable representations from data. We focused on the disentangled representation of sequential data since there is a wide range of potential applications if disentanglement representation is extended to sequential data such as video, speech, and stock market. Sequential data are characterized by dynamic and static factors: dynamic factors are time dependent, and static factors are independent of time. Previous models disentangle static and dynamic factors by explicitly modeling the priors of latent variables to distinguish between these factors. However, these models cannot disentangle representations between dynamic factors, such as disentangling "picking up" and "throwing" in robotic tasks. FAVAE can disentangle multiple dynamic factors. Since it does not require modeling priors, it can disentangle "between" dynamic factors. We conducted experiments to show that FAVAE can extract disentangled dynamic factors.
△ Less
Submitted 30 May, 2019; v1 submitted 21 February, 2019;
originally announced February 2019.
-
Novel electronic states realized by the interplay between Li diffusion and Co$^{\rm 3+}$$/$Co$^{\rm 4+}$ charge ordering in Li$_{\rm x}$CoO$_2$
Authors:
Kiyotaka Miyoshi,
Kentaro Manami,
Ryo Sasai,
Shijo Nishigori,
Jun Takeuchi
Abstract:
Measurements of dc magnetization ($M$) and electrical resistivity ($ρ$) have been carried out as a function of temperature ($T$) for layered oxide Li$_{\rm x}$CoO$_2$ (0.51$\leq$$x$$\leq$1.0) using single crystal specimens. After slow cool of the specimens down to 10 K, both of the $M$($T$) and $ρ$($T$) curves are found to exhibit a clear anomaly due to the occurrence of Co$^{\rm 3+}$$/$Co…
▽ More
Measurements of dc magnetization ($M$) and electrical resistivity ($ρ$) have been carried out as a function of temperature ($T$) for layered oxide Li$_{\rm x}$CoO$_2$ (0.51$\leq$$x$$\leq$1.0) using single crystal specimens. After slow cool of the specimens down to 10 K, both of the $M$($T$) and $ρ$($T$) curves are found to exhibit a clear anomaly due to the occurrence of Co$^{\rm 3+}$$/$Co$^{\rm 4+}$ charge ordering (CO) at $T_{\rm S}$$\sim$155 K for 0.6$\leq$$x$$\leq$0.98 (at $T_{\rm S}$$\sim$180$-$190 K for 0.5$\leq$$x$$\leq$0.55). After rapid cool of the specimens, additional anomalies are observed related to the onset of Li diffusion at $T_{\rm F1}$$\sim$370 K and$/$or $T_{\rm F2}$=120$-$130 K. Due to phase mixing with compositions of nearly LiCoO$_2$ and Li$_{\rm 2/3}$CoO$_2$, the specimens with 0.7$\lesssim$$x$$\lesssim$0.9 show anomlies both at $T_{\rm F1}$ and $T_{\rm F2}$. For 0.6$\lesssim$$x$$\lesssim$0.9, the resistivity measured after rapid cool is found to be fairly larger than that measured after slow cool below $T_{\rm S}$. The enhanced resistivity can be explained by the scenario that disordered Co$^{\rm 3+}$$/$Co$^{\rm 4+}$ arrangements, which have been observed and revealed to have an insulating electronic structure contrasting to the regular CO state in the previous scanning tunneling microscopy measurements [Phys. Rev. Lett. {\bf 111}, 126104 (2013)], are realized due to the formation of an amorphous-like structure of Li ions after rapid cool via the interlayer Coulomb coupling. An electronic phase diagram for 0.5$\leq$$x$$\leq$1.0 is proposed.
△ Less
Submitted 1 November, 2018; v1 submitted 24 August, 2018;
originally announced August 2018.
-
Enhanced Superconductivity on the Tetragonal Lattice in FeSe under Hydrostatic Pressure
Authors:
Kiyotaka Miyoshi,
Koh Morishita,
Eriko Mutou,
Masatoshi Kondo,
Osamu Seida,
Kenji Fujiwara,
Jun Takeuchi,
Shijo Nishigori
Abstract:
Superconductivity under pressure in FeSe ($T_{\rm c}$$\sim$7.5 K) has been investigated using single-crystal specimens through the measurements of DC magnetization and electrical resistivity. A characteristic three-step increase in $T_{\rm c}$ has been found under hydrostatic pressure up to $\sim$34 K above 7 GPa. The structural transition from a tetragonal phase to an orthorhombic phase (…
▽ More
Superconductivity under pressure in FeSe ($T_{\rm c}$$\sim$7.5 K) has been investigated using single-crystal specimens through the measurements of DC magnetization and electrical resistivity. A characteristic three-step increase in $T_{\rm c}$ has been found under hydrostatic pressure up to $\sim$34 K above 7 GPa. The structural transition from a tetragonal phase to an orthorhombic phase ($T_{\rm s}$$\sim$87 K) is found to disappear at $P$$\sim$2.3 GPa, above which $T_{\rm c}$ increases rapidly, suggesting that the superconductivity is enhanced by the tetragonal environment. Under non-hydrostatic pressure, the increase in $T_{\rm c}$ is suppressed and the superconductive volume fraction is considerably reduced above 2 GPa, probably owing to the breaking of the tetragonal lattice symmetry by the uniaxial stress. The intimate correlation between the enhanced (suppressed) superconductivity and the tetragonality (orthorhombicity) in the phase diagram is a common feature of FeSe and other iron-pnictide superconductors.
△ Less
Submitted 18 November, 2013; v1 submitted 28 October, 2013;
originally announced October 2013.
-
Superconductivity under pressure in $R$FeAsO$_{\rm 1-x}$F$_{\rm x}$ ($R$=La, Ce$-$Sm) by dc magnetization measurements
Authors:
Kiyotaka Miyoshi,
Eiko Kojima,
Saki Ogawa,
Yuta Shimojo,
Jun Takeuchi
Abstract:
Superconducting transition under pressure ($P$) has been investigated for optimum-doped $R$FeAsO$_{\rm 1-x}$F$_{\rm x}$ ($R$=La, Ce$-$Sm) by dc magnetization measurements. For $R$=La, $T_{\rm c}$ is found to be pressure independent up to $P$$\sim$3.0 GPa and then shows a monotonic decrease with increasing pressure. The plateau width decreases for the system with smaller lattice constants, and shri…
▽ More
Superconducting transition under pressure ($P$) has been investigated for optimum-doped $R$FeAsO$_{\rm 1-x}$F$_{\rm x}$ ($R$=La, Ce$-$Sm) by dc magnetization measurements. For $R$=La, $T_{\rm c}$ is found to be pressure independent up to $P$$\sim$3.0 GPa and then shows a monotonic decrease with increasing pressure. The plateau width decreases for the system with smaller lattice constants, and shrinks to almost zero for $R$=Sm. From the $T_{\rm c}$($P$) data, we construct the $T_{\rm c}$ evolution map on the height of the As atom from the Fe-plane $h_{\rm As}$ versus lattice constant $a$ or $c$ plane. It is shown that the characteristic $T_{\rm c}$ variations under pressure for all compounds and the rapid decrease in $T_{\rm c}$ induced by changing $R$-element from Sm to La can be described by a common $T_{\rm c}$($h_{\rm As}$, $a$ or $c$) surface, suggesting that $T_{\rm c}$ is determined by $h_{\rm As}$ and lattice constant. It is also suggested that there exists an upper limit of the lattice constant $a_{\rm ulm}$ (or $c_{\rm ulm}$), above which $dT_{\rm c}$$da$ (or $dT_{\rm c}$$dc$) changes the sign from positive to negative.
△ Less
Submitted 11 June, 2013; v1 submitted 4 March, 2013;
originally announced March 2013.
-
Magnetic and Electronic Properties of Li$_x$CoO$_2$ Single Crystals
Authors:
K. Miyoshi,
C. Iwai,
H. Kondo,
M. Miura,
S. Nishigori,
J. Takeuchi
Abstract:
Measurements of electrical resistivity ($ρ$), DC magnetization ($M$) and specific heat ($C$) have been performed on layered oxide Li$_x$CoO$_2$ (0.25$\leq$$x$$\leq$0.99) using single crystal specimens. The $ρ$ versus temperature ($T$) curve for $x$=0.90 and 0.99 is found to be insulating but a metallic behavior is observed for 0.25$\leq$$x$$\leq$0.71. At $T_{\rm S}$$\sim$155 K, a sharp anomaly is…
▽ More
Measurements of electrical resistivity ($ρ$), DC magnetization ($M$) and specific heat ($C$) have been performed on layered oxide Li$_x$CoO$_2$ (0.25$\leq$$x$$\leq$0.99) using single crystal specimens. The $ρ$ versus temperature ($T$) curve for $x$=0.90 and 0.99 is found to be insulating but a metallic behavior is observed for 0.25$\leq$$x$$\leq$0.71. At $T_{\rm S}$$\sim$155 K, a sharp anomaly is observed in the $ρ$$-$$T$, $M$$-$$T$ and $C$$/$$T$$-$$T$ curves for $x$=0.66 with thermal hysteresis, indicating the first-order charactor of the transition. The transition at $T_{\rm S}$$\sim$155 K is observed for the wide range of $x$=0.46$-$0.71. It is found that the $M$$-$$T$ curve measured after rapid cool becomes different from that after slow cool below $T_{\rm F}$, which is $\sim$130 K for $x$=0.46$-$0.71. $T_{\rm F}$ is found to agree with the temperature at which the motional narrowing in the $^7$Li NMR line width is observed, indicating that the Li ions stop diffusing and order at the regular site below $T_{\rm F}$. The ordering of Li ions below $T_{\rm F}$$\sim$130 K is likely to be triggered and stabilized by the charge ordering in CoO$_2$ layers below $T_{\rm S}$.
△ Less
Submitted 10 August, 2010; v1 submitted 24 May, 2010;
originally announced May 2010.
-
Anomalous Pressure Dependence of the Superconducting Transition Temperature in FeSe1-x Studied by DC Magnetic Measurements
Authors:
K. Miyoshi,
Y. Takaichi,
E. Mutou,
K. Fujiwara,
J. Takeuchi
Abstract:
The pressure dependence of superconducting transition temperature $T_{\rm c}$ has been investigated through the DC magnetic measurements for FeSe$_{0.8}$ and FeSe$_{1.0}$. For both samples, with increasing pressure $P$, the $T_{\rm c}$$-$$P$ curve exhibits a two-step increase, showing a local maximum of $\sim$11 K at $P$$\sim$1.0 GPa and a rapid increase with an extremely large pressure coeffici…
▽ More
The pressure dependence of superconducting transition temperature $T_{\rm c}$ has been investigated through the DC magnetic measurements for FeSe$_{0.8}$ and FeSe$_{1.0}$. For both samples, with increasing pressure $P$, the $T_{\rm c}$$-$$P$ curve exhibits a two-step increase, showing a local maximum of $\sim$11 K at $P$$\sim$1.0 GPa and a rapid increase with an extremely large pressure coefficient for $P$$>$1.5 GPa. $T_{\rm c}$ saturates at $\sim$25 K (21 K) in FeSe$_{1.0}$ (FeSe$_{0.8}$) for $P$$>$3 GPa. A rapid decrease in superconducting volume fraction is observed with an increase in $T_{\rm c}$ above 1.5 GPa, suggesting the presence of electronic inhomogeneity.
△ Less
Submitted 4 August, 2009; v1 submitted 8 June, 2009;
originally announced June 2009.
-
Superconducting Transition in the $β$-Pyrochlore AOs$_2$O$_6$ (A=Cs, Rb, K) under Pressure
Authors:
K. Miyoshi,
Y. Takaichi,
Y. Takamatsu,
M. Miura,
J. Takeuchi
Abstract:
Pressure dependence of superconducting transition temperature $T_{\rm c}$ has been determined through the DC magnetic measurements under pressure up to $P$=10 GPa for $β$-pyrochlore oxides AOs$_2$O$_6$ with A=Cs ($T_{\rm c}$=3.3 K), Rb (6.3 K) and K (9.6 K). Both for A=Rb and Cs, $T_{\rm c}$ increases with increasing $P$ and shows a saturation at $T_{\rm cm}$$\sim$8.8 K, which is considered as t…
▽ More
Pressure dependence of superconducting transition temperature $T_{\rm c}$ has been determined through the DC magnetic measurements under pressure up to $P$=10 GPa for $β$-pyrochlore oxides AOs$_2$O$_6$ with A=Cs ($T_{\rm c}$=3.3 K), Rb (6.3 K) and K (9.6 K). Both for A=Rb and Cs, $T_{\rm c}$ increases with increasing $P$ and shows a saturation at $T_{\rm cm}$$\sim$8.8 K, which is considered as the upper limit of $T_{\rm c}$ inherent in AOs$_2$O$_6$. In contrast, the $T_{\rm c}-P$ curve for KOs$_2$O$_6$ shows a sharp maximum of $\sim$10 K at $P$$\sim$0.5 GPa, and $T_{\rm c}$ is higher than $T_{\rm cm}$ for 0$\leq$$P$$\leq$1.5GPa, suggesting the enhanced superconductivity due to the rattling of K ions.
△ Less
Submitted 8 February, 2008;
originally announced February 2008.
-
Mass-Enhanced Fermi Liquid Ground State in Na$_{1.5}$Co$_2$O$_4$
Authors:
K. Miyoshi,
E. Morikuni,
K. Fujiwara,
J. Takeuchi,
T. Hamasaki
Abstract:
Magnetic, transport, and specific heat measurements have been performed on layered metallic oxide Na$_{1.5}$Co$_2$O$_4$ as a function of temperature $T$. Below a characteristic temperature $T^*$=30$-$40 K, electrical resistivity shows a metallic conductivity with a $T^2$ behavior and magnetic susceptibility deviates from the Curie-Weiss behavior showing a broad peak at $\sim$14 K. The electronic…
▽ More
Magnetic, transport, and specific heat measurements have been performed on layered metallic oxide Na$_{1.5}$Co$_2$O$_4$ as a function of temperature $T$. Below a characteristic temperature $T^*$=30$-$40 K, electrical resistivity shows a metallic conductivity with a $T^2$ behavior and magnetic susceptibility deviates from the Curie-Weiss behavior showing a broad peak at $\sim$14 K. The electronic specific heat coefficient $γ$ is $\sim$60 mJ/molK$^2$ at 2 K. No evidence for magnetic ordering is found. These behaviors suggest the formation of mass-enhanced Fermi liquid ground state analogous to that in $d$-electron heavy fermion compound LiV$_2$O$_4$.
△ Less
Submitted 1 March, 2004;
originally announced March 2004.
