arXiv subjects

Ichiro Takeuchi

Publications and source records attributed to Ichiro Takeuchi.

At least 145 records · Page 8Linked to original sources

arXiv2019

Computing Valid p-values for Image Segmentation by Selective Inference

Image segmentation is one of the most fundamental tasks of computer vision. In many practical applications, it is essential to properly evaluate the reliability of individual segmentation results. In this study, we propose a novel framework to provide the statistical significance of segmentation results in the form of p-values. Specifically, we consider a statistical hypothesis test for determining the difference between the object and the background regions. This problem is challenging because the difference can be deceptively large (called segmentation bias) due to the adaptation of the segmentation algorithm to the data. To overcome this difficulty, we introduce a statistical approach called selective inference, and develop a framework to compute valid p-values in which the segmentation bias is properly accounted for. Although the proposed framework is potentially applicable to various segmentation algorithms, we focus in this paper on graph cut-based and threshold-based segmentation algorithms, and develop two specific methods to compute valid p-values for the segmentation results obtained by these algorithms. We prove the theoretical validity of these two methods and demonstrate their practicality by applying them to segmentation problems for medical images.

cs.CV↗

arXiv2019

Bayesian Active Learning for Structured Output Design

In this paper, we propose an active learning method for an inverse problem that aims to find an input that achieves a desired structured-output. The proposed method provides new acquisition functions for minimizing the error between the desired structured-output and the prediction of a Gaussian process model, by effectively incorporating the correlation between multiple outputs of the underlying multi-valued black box output functions. The effectiveness of the proposed method is verified by applying it to two synthetic shape search problem and real data. In the real data experiment, we tackle the input parameter search which achieves the desired crystal growth rate in silicon carbide (SiC) crystal growth modeling, that is a problem of materials informatics.

stat.ML↗

arXiv2019

Computing Full Conformal Prediction Set with Approximate Homotopy

If you are predicting the label $y$ of a new object with $\hat y$, how confident are you that $y = \hat y$? Conformal prediction methods provide an elegant framework for answering such question by building a $100 (1 - α)\%$ confidence region without assumptions on the distribution of the data. It is based on a refitting procedure that parses all the possibilities for $y$ to select the most likely ones. Although providing strong coverage guarantees, conformal set is impractical to compute exactly for many regression problems. We propose efficient algorithms to compute conformal prediction set using approximated solution of (convex) regularized empirical risk minimization. Our approaches rely on a new homotopy continuation technique for tracking the solution path with respect to sequential changes of the observations. We also provide a detailed analysis quantifying its complexity.

stat.ML↗

arXiv2019

Bayesian Experimental Design for Finding Reliable Level Set under Input Uncertainty

In the manufacturing industry, it is often necessary to repeat expensive operational testing of machine in order to identify the range of input conditions under which the machine operates properly. Since it is often difficult to accurately control the input conditions during the actual usage of the machine, there is a need to guarantee the performance of the machine after properly incorporating the possible variation in input conditions. In this paper, we formulate this practical manufacturing scenario as an Input Uncertain Reliable Level Set Estimation (IU-rLSE) problem, and provide an efficient algorithm for solving it. The goal of IU-rLSE is to identify the input range in which the outputs smaller/greater than a desired threshold can be obtained with high probability when the input uncertainty is properly taken into consideration. We propose an active learning method to solve the IU-rLSE problem efficiently, theoretically analyze its accuracy and convergence, and illustrate its empirical performance through numerical experiments on artificial and real data.

stat.ML↗

arXiv2019

Microwave Meissner Screening of Proximity coupled Topological Insulator / Superconductor Bilayers

The proximity coupled topological insulator / superconductor (TI/SC) bilayer system is a representative system to realize topological superconductivity. In order to better understand this unique state and design devices from the TI/SC bilayer, a comprehensive understanding of the microscopic properties of the bilayer is required. In this work, a microwave Meissner screening study, which exploits a high-precision microwave resonator technique, is conducted on the SmB6/YB6 thin film bilayers as an example TI/SC system. The study reveals spatially dependent electrodynamic screening response of the TI/SC system that is not accessible to other techniques, from which the corresponding microscopic properties of a TI/SC bilayer can be obtained. The TI thickness dependence of the effective penetration depth suggests the existence of a bulk insulating region in the TI layer. The spatially dependent electrodynamic screening model analysis provides an estimate for the characteristic lengths of the TI/SC bilayer: normal penetration depth, normal coherence length, and the thickness of the surface states. We also discuss implications of these characteristic lengths on the design of a vortex Majorana device such as the radius of the vortex core, the energy splitting due to intervortex tunneling, and the minimum thickness required for a device.

cond-mat.supr-con↗

arXiv2019

Fatigue-resistant high-performance elastocaloric materials via additive manufacturing

Elastocaloric cooling, which exploits the latent heat released and absorbed as stress-induced phase transformations are reversibly cycled in shape memory alloys, has recently emerged as a frontrunner in non-vapor-compression cooling technologies. The intrinsically high thermodynamic efficiency of elastocaloric materials is limited only by work hysteresis. Here, we report on creating high-performance low-hysteresis elastocaloric cooling materials via additive manufacturing of Titanium-Nickel (Ti-Ni) alloys. Contrary to established knowledge of the physical metallurgy of Ti-Ni alloys, intermetallic phases are found to be beneficial to elastocaloric performances when they are combined with the binary Ti-Ni compound in nanocomposite configurations. The resulting microstructure gives rise to quasi-linear stress-strain behaviors with extremely small hysteresis, leading to enhancement in the materials efficiency by a factor of five. Furthermore, despite being composed of more than 50% intermetallic phases, the reversible, repeatable elastocaloric performance of this material is shown to be stable over one million cycles. This result opens the door for direct implementation of additive manufacturing to elastocaloric cooling systems where versatile design strategy enables both topology optimization of heat exchangers as well as unique microstructural control of metallic refrigerants.

cond-mat.mtrl-sci↗

arXiv2019

Selective Inference via Marginal Screening for High Dimensional Classification

Post-selection inference is a statistical technique for determining salient variables after model or variable selection. Recently, selective inference, a kind of post-selection inference framework, has garnered the attention in the statistics and machine learning communities. By conditioning on a specific variable selection procedure, selective inference can properly control for so-called selective type I error, which is a type I error conditional on a variable selection procedure, without imposing excessive additional computational costs. While selective inference can provide a valid hypothesis testing procedure, the main focus has hitherto been on Gaussian linear regression models. In this paper, we develop a selective inference framework for binary classification problem. We consider a logistic regression model after variable selection based on marginal screening, and derive the high dimensional statistical behavior of the post-selection estimator. This enables us to asymptotically control for selective type I error for the purposes of hypothesis testing after variable selection. We conduct several simulation studies to confirm the statistical power of the test, and compare our proposed method with data splitting and other methods.

stat.ME↗

arXiv2019

Safe Grid Search with Optimal Complexity

Popular machine learning estimators involve regularization parameters that can be challenging to tune, and standard strategies rely on grid search for this task. In this paper, we revisit the techniques of approximating the regularization path up to predefined tolerance $ε$ in a unified framework and show that its complexity is $O(1/\sqrt[d]ε)$ for uniformly convex loss of order $d \geq 2$ and $O(1/\sqrtε)$ for Generalized Self-Concordant functions. This framework encompasses least-squares but also logistic regression, a case that as far as we know was not handled as precisely in previous works. We leverage our technique to provide refined bounds on the validation error as well as a practical algorithm for hyperparameter tuning. The latter has global convergence guarantee when targeting a prescribed accuracy on the validation set. Last but not least, our approach helps relieving the practitioner from the (often neglected) task of selecting a stopping criterion when optimizing over the training set: our method automatically calibrates this criterion based on the targeted accuracy on the validation set.

stat.ML↗

arXiv2019

Statistically Discriminative Sub-trajectory Mining

We study the problem of discriminative sub-trajectory mining. Given two groups of trajectories, the goal of this problem is to extract moving patterns in the form of sub-trajectories which are more similar to sub-trajectories of one group and less similar to those of the other. We propose a new method called Statistically Discriminative Sub-trajectory Mining (SDSM) for this problem. An advantage of the SDSM method is that the statistical significance of the extracted sub-trajectories are properly controlled in the sense that the probability of finding a false positive sub-trajectory is smaller than a specified significance threshold alpha (e.g., 0.05), which is indispensable when the method is used in scientific or social studies under noisy environment. Finding such statistically discriminative sub-trajectories from massive trajectory dataset is both computationally and statistically challenging. In the SDSM method, we resolve the difficulties by introducing a tree representation among sub-trajectories and running an efficient permutation-based statistical inference method on the tree. To the best of our knowledge, SDSM is the first method that can efficiently extract statistically discriminative sub-trajectories from massive trajectory dataset. We illustrate the effectiveness and scalability of the SDSM method by applying it to a real-world dataset with 1,000,000 trajectories which contains 16,723,602,505 sub-trajectories.

stat.ML↗

arXiv2019

High-throughput investigation of tunable superconductivity in FeSe films

There is an ongoing debate about the relative importance of structural change versus doping charge carriers on the mechanism of superconductivity in Fe-based materials. Elucidating this issue is a major challenge since it would require a large number of samples where structure properties or the carrier density is systematically varied. FeSe, with its structural simplicity, is an ideal platform for addressing this question. It has been demonstrated that the superconductivity in this material can be controlled through crystal lattice tuning, as well as electronic structure manipulation. Here, we apply a high-throughput methodology to FeSe to systematically delineate the interdependence of its structural and electronic properties. Using a dual-beam pulsed laser deposition, we have generated FeSe films with a marked gradient in the superconducting transition temperature (below 2 K < Tc < 12 K) across 1 cm width of the films. The Tc gradient films display ~ 1% continuous stretch and compression in the out-of-plane and in-plane lattice constants respectively, triggering the continuous enhancement of superconductivity. Combining transport and angular-resolved photoemission measurements on uniform FeSe films with tunable Tc from 3 K to 14 K, we find that the electron carrier density is intimately correlated with Tc, i.e., it increases by a factor of 6 and ultimately surpasses the almost constant hole concentration. Our transmission electron microscope and band structure calculations reveal that rather than by shifting the chemical potential, the enhanced superconductivity is linked to the selective adjustment of the dxy band dispersion across the Fermi level by means of reduced local lattice distortions. Therefore, such novel mechanism provides a key to understand discrete superconducting phases in FeSe.

cond-mat.supr-con↗

arXiv2019

Active learning for enumerating local minima based on Gaussian process derivatives

We study active learning (AL) based on Gaussian Processes (GPs) for efficiently enumerating all of the local minimum solutions of a black-box function. This problem is challenging due to the fact that local solutions are characterized by their zero gradient and positive-definite Hessian properties, but those derivatives cannot be directly observed. We propose a new AL method in which the input points are sequentially selected such that the confidence intervals of the GP derivatives are effectively updated for enumerating local minimum solutions. We theoretically analyze the proposed method and demonstrate its usefulness through numerical experiments.

stat.ML↗

arXiv2019

Materials development by interpretable machine learning

Machine learning technologies are expected to be great tools for scientific discoveries. In particular, materials development (which has brought a lot of innovation by finding new and better functional materials) is one of the most attractive scientific fields. To apply machine learning to actual materials development, collaboration between scientists and machine learning is becoming inevitable. However, such collaboration has been restricted so far due to black box machine learning, in which it is difficult for scientists to interpret the data-driven model from the viewpoint of material science and physics. Here, we show a material development success story that was achieved by good collaboration between scientists and one type of interpretable (explainable) machine learning called factorized asymptotic Bayesian inference hierarchical mixture of experts (FAB/HMEs). Based on material science and physics, we interpreted the data-driven model constructed by the FAB/HMEs, so that we discovered surprising correlation and knowledge about thermoelectric material. Guided by this, we carried out actual material synthesis that led to identification of a novel spin-driven thermoelectric material with the largest thermopower to date.

cond-mat.mtrl-sci↗

arXiv2019

Observation of perfect Andreev reflection due to Klein paradox in a topological superconducting state

In 1928, P. Dirac proposed a new wave equation to describe relativistic electrons. Shortly afterwards, O. Klein solved a simple potential step problem for the Dirac equation and stumbled upon an apparent paradox - the potential becomes transparent when the height is larger than the electron energy. For massless particles, backscattering is completely forbidden in Klein tunneling, leading to perfect transmission through any potential barrier. Recent advent of condensed matter systems with Dirac-like excitations, such as graphene and topological insulators (TIs), has opened the possibility of observing the Klein tunneling experimentally. In the surface states of TIs, fermions are bound by spin-momentum locking, and are thus immune to backscattering due to time-reversal symmetry. Here we report the observation of perfect Andreev reflection in point contact spectroscopy - a clear signature of Klein tunneling and a manifestation of the underlying relativistic physics of a proximity-induced superconducting state in a topological Kondo insulator.

cond-mat.mes-hall↗

arXiv2018

Knowledge-Transfer based Cost-effective Search for Interface Structures: A Case Study on fcc-Al [110] Tilt Grain Boundary

Determining the atomic configuration of an interface is one of the most important issues in materials science research. Although theoretical simulations are effective tools, an exhaustive search is computationally prohibitive due to the high degrees of freedom of the interface structure. In the interface structure search, multiple energy surfaces created by a variety of orientation angles need to be explored, and the necessary computational costs for different angles vary substantially owing to significant variations in the supercell sizes. In this paper, we introduce two machine-learning concepts, called transfer learning and cost-sensitive search, to the interface-structure search. As a case study, we demonstrate the effectiveness of our method, called cost-sensitive multi-task Bayesian optimization (CMB), using the fcc-Al [110] tilt grain boundary. Four microscopic parameters, the three-dimensional rigid body translation, and the number of atomic columns, are optimized by transferring knowledge of energy surfaces among different orientation angles. We show that transferring knowledge of different energy surfaces can accelerate the structure search, and that considering the cost variations further improves the total efficiency.

cond-mat.mtrl-sci↗

arXiv2018

Safe Triplet Screening for Distance Metric Learning

We study safe screening for metric learning. Distance metric learning can optimize a metric over a set of triplets, each one of which is defined by a pair of same class instances and an instance in a different class. However, the number of possible triplets is quite huge even for a small dataset. Our safe triplet screening identifies triplets which can be safely removed from the optimization problem without losing the optimality. Compared with existing safe screening studies, triplet screening is particularly significant because of (1) the huge number of possible triplets, and (2) the semi-definite constraint in the optimization. We derive several variants of screening rules, and analyze their relationships. Numerical experiments on benchmark datasets demonstrate the effectiveness of safe triplet screening.

stat.ML↗

arXiv2018

Selecting the Best in GANs Family: a Post Selection Inference Framework

"Which Generative Adversarial Networks (GANs) generates the most plausible images?" has been a frequently asked question among researchers. To address this problem, we first propose an \emph{incomplete} U-statistics estimate of maximum mean discrepancy $\mathrm{MMD}_{inc}$ to measure the distribution discrepancy between generated and real images. $\mathrm{MMD}_{inc}$ enjoys the advantages of asymptotic normality, computation efficiency, and model agnosticity. We then propose a GANs analysis framework to select and test the "best" member in GANs family using the Post Selection Inference (PSI) with $\mathrm{MMD}_{inc}$. In the experiments, we adopt the proposed framework on 7 GANs variants and compare their $\mathrm{MMD}_{inc}$ scores.

cs.LG↗

arXiv2018

AFLOW-CHULL: Cloud-oriented platform for autonomous phase stability analysis

$\textit{A priori}$ prediction of phase stability of materials is a challenging practice, requiring knowledge of all energetically-competing structures at formation conditions. Large materials repositories $\unicode{x2014}$ housing properties of both experimental and hypothetical compounds $\unicode{x2014}$ offer a path to prediction through the construction of informatics-based, $\textit{ab-initio}$ phase diagrams. However, limited access to relevant data and software infrastructure has rendered thermodynamic characterizations largely peripheral, despite their continued success in dictating synthesizability. Herein, a new module is presented for autonomous thermodynamic stability analysis implemented within the open-source, $\textit{ab-initio}$ framework AFLOW. Powered by the AFLUX Search-API, AFLOW-CHULL leverages data of more than 1.8 million compounds currently characterized in the AFLOW.org repository and can be employed locally from any UNIX-like computer. The module integrates a range of functionality: the identification of stable phases and equivalent structures, phase coexistence, measures for robust stability, and determination of decomposition reactions. As a proof-of-concept, thorough thermodynamic characterizations have been performed for more than 1,300 binary and ternary systems, enabling the identification of several candidate phases for synthesis based on their relative stability criterion $\unicode{x2014}$ including 18 promising $C15_{b}$-type structures and two half-Heuslers. In addition to a full report included herein, an interactive, online web application has been developed showcasing the results of the analysis, and is located at aflow.org/aflow-chull.

cond-mat.mtrl-sci↗

arXiv2018

Machine-learning guided discovery of a high-performance spin-driven thermoelectric material

Thermoelectric conversion using Seebeck effect for generation of electricity is becoming an indispensable technology for energy harvesting and smart thermal management. Recently, the spin-driven thermoelectric effects (STEs), which employ emerging phenomena such as the spin-Seebeck effect (SSE) and the anomalous Nernst effect (ANE), have garnered much attention as a promising path towards low cost and versatile thermoelectric technology with easily scalable manufacturing. However, progress in development of STE devices is hindered by the lack of understanding of the mechanism and materials parameters that govern the STEs. To address this problem, we enlist machine learning modeling to establish the key physical parameters controlling SSE. Guided by these models, we have carried out a high-throughput experiment which led to the identification of a novel STE material with a thermopower an order of magnitude larger than that of the current generation STE devices.

cond-mat.mtrl-sci↗