Ichiro Takeuchi

Ichiro Takeuchi
Are you Ichiro Takeuchi?

Claim your profile, edit publications, add additional information:

Contact Details

Name
Ichiro Takeuchi
Affiliation
Location

Pubs By Year

Pub Categories

 
Statistics - Machine Learning (14)
 
Physics - Materials Science (11)
 
Computer Science - Learning (8)
 
Physics - Superconductivity (8)
 
Physics - Strongly Correlated Electrons (5)
 
Physics - Mesoscopic Systems and Quantum Hall Effect (2)
 
Computer Science - Cryptography and Security (1)
 
Mathematics - Optimization and Control (1)
 
Statistics - Methodology (1)

Publications Authored By Ichiro Takeuchi

We study primal-dual type stochastic optimization algorithms with non-uniform sampling. Our main theoretical contribution in this paper is to present a convergence analysis of Stochastic Primal Dual Coordinate (SPDC) Method with arbitrary sampling. Based on this theoretical framework, we propose Optimality Violation-based Sampling SPDC (ovsSPDC), a non-uniform sampling method based on Optimality Violation. Read More

We propose a novel kernel based post selection inference (PSI) algorithm, which can not only handle non-linearity in data but also structured output such as multi-dimensional and multi-label outputs. Specifically, we develop a PSI algorithm for independence measures, and propose the Hilbert-Schmidt Independence Criterion (HSIC) based PSI algorithm (hsicInf). The novelty of the proposed algorithm is that it can handle non-linearity and/or structured data through kernels. Read More

We study large-scale classification problems in changing environments where a small part of the dataset is modified, and the effect of the data modification must be quickly incorporated into the classifier. When the entire dataset is large, even if the amount of the data modification is fairly small, the computational cost of re-training the classifier would be prohibitively large. In this paper, we propose a novel method for efficiently incorporating such a data modification effect into the classifier without actually re-training it. Read More

The proximity effect at the interface between a topological insulator (TI) and a superconductor is predicted to give rise to chiral topological superconductivity and Majorana fermion excitations. In most TIs studied to date, however, the conducting bulk states have overwhelmed the transport properties and precluded the investigation of the interplay of the topological surface state and Cooper pairs. Here, we demonstrate the superconducting proximity effect in the surface state of SmB6 thin films which display bulk insulation at low temperatures. Read More

In 2006, a novel cobalt-based superalloy was discovered [1] with mechanical properties better than some conventional nickel-based superalloys. As with conventional superalloys, its high performance arises from the precipitate-hardening effect of a coherent L1$_2$ phase, which is in two-phase equilibrium with the fcc matrix. Inspired by this unexpected discovery of an L1$_2$ ternary phase, we performed a first-principles search through 2224 ternary metallic systems for analogous precipitate-hardening phases of the form $X_{3}$[$A_{0. Read More

Privacy concern has been increasingly important in many machine learning (ML) problems. We study empirical risk minimization (ERM) problems under secure multi-party computation (MPC) frameworks. Main technical tools for MPC have been developed based on cryptography. Read More

Discovering statistically significant patterns from databases is an important challenging problem. The main obstacle of this problem is in the difficulty of taking into account the selection bias, i.e. Read More

In this paper we study predictive pattern mining problems where the goal is to construct a predictive model based on a subset of predictive patterns in the database. Our main contribution is to introduce a novel method called safe pattern pruning (SPP) for a class of predictive pattern mining problems. The SPP method allows us to efficiently find a superset of all the predictive patterns in the database that are needed for the optimal predictive model. Read More

The problem of learning a sparse model is conceptually interpreted as the process of identifying active features/samples and then optimizing the model over them. Recently introduced safe screening allows us to identify a part of non-active features/samples. So far, safe screening has been individually studied either for feature screening or for sample screening. Read More

In this paper, we propose a selective sampling procedure to preferentially evaluate a potential energy surface (PES) in a part of the configuration space governing a physical property of interest. The proposed sampling procedure is based on a machine learning method called the Gaussian process (GP), which is used to construct a statistical model of the PES for identifying the region of interest in the configuration space. We demonstrate the efficacy of the proposed procedure for atomic diffusion and ionic conduction, specifically the proton conduction in a well-studied proton-conducting oxide, barium zirconate BaZrO3. Read More

We utilize terahertz time domain spectroscopy to investigate thin films of the heavy fermion compound SmB6, a prototype Kondo insulator. Temperature dependent terahertz (THz) conductivity measurements reveal a rapid decrease in the Drude weight and carrier scattering rate at ~T*=20 K, well below the hybridization gap onset temperature (100 K). Moreover, a low-temperature conductivity plateau (below 20K) indicates the emergence of a surface state with an effective electron mass of 0. Read More

High-throughput ab-initio calculations, cluster expansion techniques and thermodynamic modeling have been synergistically combined to characterize the binodal and the spinodal decompositions features in the pseudo-binary lead chalcogenides PbSe-PbTe, PbS-PbTe, and PbS-PbSe. While our results agree with the available experimental data, our consolute temperatures substantially improve with respect to previous computational modeling. The computed phase diagrams corroborate that the formation of spinodal nanostructures causes low thermal conductivities in these alloys. Read More

The neural-network interatomic potential for crystalline and liquid Si has been developed using the forward stepwise regression technique to reduce the number of bases with keeping the accuracy of the potential. This approach of making the neural-network potential enables us to construct the accurate interatomic potentials with less and important bases selected systematically and less heuristically. The evaluation of bulk crystalline properties, and dynamic properties of liquid Si show good agreements between the neural-network potential and ab-initio results. Read More

In support vector machine (SVM) applications with unreliable data that contains a portion of outliers, non-robustness of SVMs often causes considerable performance deterioration. Although many approaches for improving the robustness of SVMs have been studied, two major challenges remain in robust SVM learning. First, robust learning algorithms are essentially formulated as non-convex optimization problems. Read More

Taking into account high-order interactions among covariates is valuable in many practical regression problems. This is, however, computationally challenging task because the number of high-order interaction features to be considered would be extremely large unless the number of covariates is sufficiently small. In this paper, we propose a novel efficient algorithm for LASSO-based sparse learning of such high-order interaction models. Read More

Since the discovery of n-type copper oxide superconductors, the evolution of electron- and hole-bands and its relation to the superconductivity have been seen as a key factor in unveiling the mechanism of high-Tc superconductors. So far, the occurrence of electrons and holes in n-type copper oxides has been achieved by chemical doping, pressure, and/or deoxygenation. However, the observed electronic properties are blurred by the concomitant effects such as change of lattice structure, disorder, etc. Read More

We introduce a novel sensitivity analysis framework for large scale classification problems that can be used when a small number of instances are incrementally added or removed. For quickly updating the classifier in such a situation, incremental learning algorithms have been intensively studied in the literature. Although they are much more efficient than solving the optimization problem from scratch, their computational complexity yet depends on the entire training set size. Read More

Careful tuning of a regularization parameter is indispensable in many machine learning tasks because it has a significant impact on generalization performances. Nevertheless, current practice of regularization parameter tuning is more of an art than a science, e.g. Read More

SmB6 has been predicted and verified as a prototype of topological Kondo insulators (TKIs). Here we report longitudinal magnetoresistance and Hall coefficient measurements on co-sputtered nanocrystalline SmB6 films and try to find possible signatures of their topological properties. The magnetoresistance (MR) at 2 K is positive and linear (LPMR) at low field and becomes negative and quadratic at higher field. Read More

Transition-metal oxides offer an opportunity to explore unconventional superconductors, where the superconductivity (SC) is often interrelated with novel phenomena such as spin/charge order, fluctuations, and Fermi surface instability (1-3). LiTi2O4 (LTO) is a unique compound in that it is the only known spinel oxide superconductor. In addition to electron-phonon coupling, electron-electron and spin fluctuation contributions have been suggested as playing important roles in the microscopic mechanism for its superconductivity (4-8). Read More

We report on the evolution of the magnetic structure of BiFeO3 thin films grown on SrTiO3 substrates as a function of Sm doping. We determined the magnetic structure using neutron diffraction. We found that as Sm increases, the magnetic structure evolves from a cycloid to a G-type antiferromagnet at the morphotropic phase boundary, where there is a large piezoelectric response due to an electric-field induced structural transition. Read More

Practical model building processes are often time-consuming because many different models must be trained and validated. In this paper, we introduce a novel algorithm that can be used for computing the lower and the upper bounds of model validation errors without actually training the model itself. A key idea behind our algorithm is using a side information available from a suboptimal model. Read More

Sparse classifiers such as the support vector machines (SVM) are efficient in test-phases because the classifier is characterized only by a subset of the samples called support vectors (SVs), and the rest of the samples (non SVs) have no influence on the classification result. However, the advantage of the sparsity has not been fully exploited in training phases because it is generally difficult to know which sample turns out to be SV beforehand. In this paper, we introduce a new approach called safe sample screening that enables us to identify a subset of the non-SVs and screen them out prior to the training phase. Read More

We have fabricated Fe-B thin film composition spreads in search of possible superconducting phases following a theoretical prediction by Kolmogorov et al.^1 Co-sputtering was used to deposit spreads covering a large compositional region of the Fe-B binary phase diagram. A trace of superconducting phase was found in the nanocrystalline part of the spread, where the film undergoes a metal to insulator transition as a function of composition in a region with the average composition of FeB_2. Read More

We address the problem of estimating the difference between two probability densities. A naive approach is a two-step procedure of first estimating two densities separately and then computing their difference. However, such a two-step procedure does not necessarily work well because the first step is performed without regard to the second step and thus a small error incurred in the first stage can cause a big error in the second stage. Read More

We have fabricated c-axis point contact junctions between high-quality LiFeAs single crystals and Pb or Au tips in order to study the nature of the superconducting order parameter of LiFeAs, one of the few stoichiometric iron-based superconductors. The observation of the Josephson current in c-axis junctions with a conventional s-wave superconductor as the counterelectrode indicates that the pairing symmetry in LiFeAs is not pure d-wave or pure spin-triplet p-wave. A superconducting gap is clearly observed in point contact Andreev reflection measurements performed on both Pb/LiFeAs and Au/LiFeAs junctions. Read More

We consider a suboptimal solution path algorithm for the Support Vector Machine. The solution path algorithm is an effective tool for solving a sequence of a parametrized optimization problems in machine learning. The path of the solutions provided by this algorithm are very accurate and they satisfy the optimality conditions more strictly than other SVM optimization algorithms. Read More

We report the piezoresisitivity in magnetic thin films of FeGa and their use for fabricating self transducing microcantilevers. The actuation occurs as a consequence of both the ferromagnetic and magnetostrictive property of FeGa thin films, while the deflection readout is achieved by exploiting the piezoresisitivity of these films. This self-sensing, self-actuating micromechanical system involves a very simple bilayer structure, which eliminates the need for the more complex piezoelectric stack that is commonly used in active cantilevers. Read More

An instance-weighted variant of the support vector machine (SVM) has attracted considerable attention recently since they are useful in various machine learning tasks such as non-stationary data analysis, heteroscedastic data modeling, transfer learning, learning to rank, and transduction. An important challenge in these scenarios is to overcome the computational bottleneck---instance weights often change dynamically or adaptively, and thus the weighted SVM solutions must be repeatedly computed. In this paper, we develop an algorithm that can efficiently and exactly update the weighted SVM solutions for arbitrary change of instance weights. Read More

Atomic resolution imaging is demonstrated using a hybrid scanning tunneling/near-field microwave microscope (microwave-STM). The microwave channels of the microscope correspond to the resonant frequency and quality factor of a coaxial microwave resonator, which is built in to the STM scan head and coupled to the probe tip. We find that when the tip-sample distance is within the tunneling regime, we obtain atomic resolution images using the microwave channels of the microwave-STM. Read More

We have systematically investigated the doping and the directional dependence of the gap structure in the 122-type iron pnictide superconductors by point contact Andreev reflection spectroscopy. The studies were performed on single crystals of Ba1-xKxFe2As2 (x = 0.29, 0. Read More

We have fabricated c-axis Josephson junctions on single crystals of (Ba,K)Fe2As2 by using Pb as the counter electrode in two geometries, planar and point contact. Junctions in both geometries show resistively shunted junction I-V curves below the Tc of the counter electrode. Microwave induced steps were observed in the I-V curves, and the critical currents are suppressed with an in-plane magnetic field in a manner consistent with the small junction limit. Read More