Takashi Washio

Osaka University, Suika, Ōsaka, Japan

Are you Takashi Washio?

Claim your profile

Publications (177)38.5 Total impact

  • Naoki Tanaka, Shohei Shimizu, Takashi Washio
    [Show abstract] [Hide abstract]
    ABSTRACT: A large amount of observational data has been accumulated in various fields in recent times, and there is a growing need to estimate the generating processes of these data. A linear non-Gaussian acyclic model (LiNGAM) based on the non-Gaussianity of external influences has been proposed to estimate the data-generating processes of variables. However, the results of the estimation can be biased if there are latent classes. In this paper, we first review LiNGAM, its extended model, as well as the estimation procedure for LiNGAM in a Bayesian framework. We then propose a new Bayesian estimation procedure that solves the problem.
    08/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Discovering causal relations among observed variables in a given data set is a major objective in studies of statistics and artificial intelligence. Recently, some techniques to discover a unique causal model have been explored based on non-Gaussianity of the observed data distribution. However, most of these are limited to continuous data. In this paper, we present a novel causal model for binary data and propose an efficient new approach to deriving the unique causal model governing a given binary data set under skew distributions of external binary noises. Experimental evaluation shows excellent performance for both artificial and real world data sets.
    01/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The notion of causality is used in many situations dealing with uncertainty. We consider the problem whether causality can be identified given data set generated by discrete random variables rather than continuous ones. In particular, for non-binary data, thus far it was only known that causality can be identified except rare cases. In this paper, we present necessary and sufficient condition for an integer modular acyclic additive noise (IMAN) of two variables. In addition, we relate bivariate and multivariate causal identifiability in a more explicit manner, and develop a practical algorithm to find the order of variables and their parent sets. We demonstrate its performance in applications to artificial data and real world body motion data with comparisons to conventional methods.
    01/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The accurate detection of small deviations in given density matrices is important for quantum information processing. Here we propose a new method based on the concept of data mining. We demonstrate that the proposed method can more accurately detect small erroneous deviations in reconstructed density matrices, which contain intrinsic fluctuations due to the limited number of samples, than a naive method of checking the trace distance from the average of the given density matrices. This method has the potential to be a key tool in broad areas of physics where the detection of small deviations of quantum states reconstructed using a limited number of samples are essential.
    01/2014; 89(2).
  • Jonathan R. Wells, Kai Ming Ting, Takashi Washio
    [Show abstract] [Hide abstract]
    ABSTRACT: Despite their wide spread use, nearest neighbour density estimators have two fundamental limitations: O(n2)O(n2) time complexity and O(n) space complexity. Both limitations constrain nearest neighbour density estimators to small data sets only. Recent progress using indexing schemes has improved to near linear time complexity only. We propose a new approach, called LiNearN for Linear time Nearest Neighbour algorithm, that yields the first nearest neighbour density estimator having O(n) time complexity and constant space complexity, as far as we know. This is achieved without using any indexing scheme because LiNearN uses a subsampling approach for which the subsample values are significantly less than the data size. Like existing density estimators, our asymptotic analysis reveals that the new density estimator has a parameter to trade off between bias and variance. We show that algorithms based on the new nearest neighbour density estimator can easily scale up to data sets with millions of instances in anomaly detection and clustering tasks.
    Pattern Recognition. 01/2014; 47(8):2702–2720.
  • [Show abstract] [Hide abstract]
    ABSTRACT: We often use a discrete time vector autoregressive (DVAR) model to analyse continuous time, multivariate, linear Markov systems through their time series data sampled at discrete time steps. However, the DVAR model has been considered not to be structural representation and hence not to have bijective correspondence with system dynamics in general. In this paper, we characterize the relationships of the DVAR model with its corresponding structural vector AR (SVAR) and continuous time vector AR (CVAR) models through finite difference approximation of time differentials. Our analysis shows that the DVAR model of a continuous time, multivariate, linear Markov system bijectively corresponds to the system dynamics. Further we clarify that the SVAR and the CVAR models are uniquely reproduced from their DVAR model under a highly generic condition. Based on these results, we propose a novel Continuous time and Structural Vector AutoRegressive (CSVAR) modeling approach for continuous time, linear Markov systems to derive the SVAR and the CVAR models from their DVAR model empirically derived from the observed time series. We demonstrate its superior performance through some numerical experiments on both artificial and real world data.
    Proceedings of the 2013 IEEE 13th International Conference on Data Mining Workshops; 12/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We consider learning a causal ordering of variables in a linear nongaussian acyclic model called LiNGAM. Several methods have been shown to consistently estimate a causal ordering assuming that all the model assumptions are correct. But the estimation results could be distorted if some assumptions are violated. In this letter, we propose a new algorithm for learning causal orders that is robust against one typical violation of the model assumptions: latent confounders. The key idea is to detect latent confounders by testing independence between estimated external influences and find subsets (parcels) that include variables unaffected by latent confounders. We demonstrate the effectiveness of our method using artificial data and simulated brain imaging data.
    Neural Computation 10/2013; · 1.76 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The analysis of multimedia application traces can reveal important information to enhance program execution comprehension. However typical size of traces can be in gigabytes, which hinders their effective exploitation by application developers. In this paper, we study the problem of finding a set of sequences of events that allows a reduced-size rewriting of the original trace. These sequences of events, that we call blocks, can simplify the exploration of large execution traces by allowing application developers to see an abstraction instead of low-level events. The problem of computing such set of blocks is NP-hard and naive approaches lead to prohibitive running times that prevent analysing real world traces. We propose a novel algorithm that directly mines the set of blocks. Our experiments show that our algorithm can analyse real traces of up to two hours of video. We also show experimentally the quality of the set of blocks proposed, and the interest of the rewriting to understand actual trace data.
    Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining; 08/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: The accuracy of active learning is critically influenced by the existence of noisy labels given by a noisy oracle. In this paper, we propose a novel pool-based active learning framework through robust measures based on density power divergence. By minimizing density power divergence, such as β-divergence and γ-divergence, one can estimate the model accurately even under the existence of noisy labels within data. Accordingly, we develop query selecting measures for pool-based active learning using these divergences. In addition, we propose an evaluation scheme for these measures based on asymptotic statistical analyses, which enables us to perform active learning by evaluating an estimation error directly. Experiments with benchmark datasets and real-world image datasets show that our active learning scheme performs better than several baseline methods.
    Neural networks: the official journal of the International Neural Network Society 05/2013; 46C:133-143. · 1.88 Impact Factor
  • Transactions of the Japanese Society for Artificial Intelligence 01/2013; 28(1):13-21.
  • [Show abstract] [Hide abstract]
    ABSTRACT: The prognoses for patients with certain diseases are estimated by averaging the results of clinical trials. To investigate the possibility of deriving a mathematical formula for the estimation of prognosis, we formulated the equation τ=f(x(1), …, x(p)), where x(1), …, x(p) are clinical features and τ represents the clinical outcome for heart failure (HF). We attempted to determine the function to mathematically formulate the relationship between clinical features and outcomes for these patients. We followed 151 patients (mean age: 68.6±14.6 years; men: 61.6%) who were consecutively hospitalized and discharged as a result of acute decompensated HF (ADHF) between May 2006 and December 2009. The mathematical analysis was performed through a probabilistic modeling of the relational data by assuming a Poisson process for rehospitalization owing to HF and by linearly approximating the relationship between the clinical factors and the mean elapsed time to rehospitalization. The former assumption was validated by a statistical test of the data, and the contribution of each parameter was assessed based on the coefficients of the linear relation. Using a regularization method to analyze 402 clinical parameters, we identified 252 factors that substantially influenced the elapsed time until rehospitalization. With the probability model based on the Poisson process, the actual (X; 388±377 days) and estimated (Y; 398±381 days) elapsed times to rehospitalization were tightly correlated (Y=1.0076X+6.5531, R(2)=0.9879, P<0.0001). We established a mathematical formula that closely predicts the clinical outcomes of patients who are hospitalized with ADHF and discharged after appropriate treatment.Hypertension Research advance online publication, 20 December 2012; doi:10.1038/hr.2012.200.
    Hypertension Research 12/2012; · 2.79 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The analysis of multimedia application traces can reveal important information to enhance program comprehen- sion. However traces can be very large, which hinders their effective exploitation. In this paper, we study the problem of finding a k-golden set of blocks that best characterize data. Sequential pattern mining can help to automatically discover the blocks, and we called k-golden set,asetof k blocks that maximally covers the trace. These kind of blocks can simplify the exploration of large traces by allowing programmers to see an abstraction instead of low-level events. We propose an approach for mining golden blocks and finding coverage of frames. The experiments carried out on video and audio application decoding show very promising results.
    12/2012;
  • Source
    Satoshi Hara, Takashi Washio
    [Show abstract] [Hide abstract]
    ABSTRACT: Properties of data are frequently seen to vary depending on the sampled situations, which usually change along a time evolution or owing to environmental effects. One way to analyze such data is to find invariances, or representative features kept constant over changes. The aim of this paper is to identify one such feature, namely interactions or dependencies among variables that are common across multiple datasets collected under different conditions. To that end, we propose a common substructure learning (CSSL) framework based on a graphical Gaussian model. We further present a simple learning algorithm based on the Dual Augmented Lagrangian and the Alternating Direction Method of Multipliers. We confirm the performance of CSSL over other existing techniques in finding unchanging dependency structures in multiple datasets through numerical simulations on synthetic data and through a real world application to anomaly detection in automobile sensors.
    Neural networks: the official journal of the International Neural Network Society 11/2012; 38C:23-38. · 1.88 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The performance of active learning (AL) is crucially influenced by the existence of outliers in input samples. In this paper, we propose a robust pool-based AL measure based on the density power divergence. It is known that the density power divergence can be accurately estimated even under the existence of outliers within data. We further derive an AL scheme based on an asymptotic statistical analysis on the M-estimator. The performance of the proposed framework is investigated empirically using artificial and real-world data.
    Proceedings of the 19th international conference on Neural Information Processing - Volume Part III; 11/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Methods for mining graph sequences have recently attracted considerable interest from researchers in the data-mining field. A graph sequence is one of the data structures that represent changing networks. The objective of graph sequence mining is to enumerate common changing patterns appearing more frequently than a given threshold from graph sequences. Syntactic dependency analysis has been recognized as a basic process in natural language processing. In a transition-based parser for dependency analysis, a transition sequence can be represented by a graph sequence where each graph, vertex, and edge respectively correspond to a state, word, and dependency. In this paper, we propose a method for mining rules for rewriting states reaching incorrect final states to states reaching the correct final state, and propose a dependency parser that uses rewriting rules. The proposed parser is comparable to conventional dependency parsers in terms of computational complexity.
    Proceedings of the 12th Pacific Rim international conference on Trends in Artificial Intelligence; 09/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Non-stationary effects are ubiquitous in real world data. In many settings, the observed signals are a mixture of underlying stationary and non-stationary sources that cannot be measured directly. For example, in EEG analysis, electrodes on the scalp record the activity from several sources located inside the brain, which one could only measure invasively. Discerning stationary and non-stationary contributions is an important step towards uncovering the mechanisms of the data generating system. To that end, in Stationary Subspace Analysis (SSA), the observed signal is modeled as a linear superposition of stationary and non-stationary sources, where the aim is to separate the two groups in the mixture. In this paper, we propose the first SSA algorithm that has a closed form solution. The novel method, Analytic SSA (ASSA), is more than 100 times faster than the state-of-the-art, numerically stable, and guaranteed to be optimal when the covariance between stationary and non-stationary sources is time-constant. In numerical simulations on wide range of settings, we show that our method yields superior results, even for signals with time-varying group-wise covariance. In an application to geophysical data analysis, ASSA extracts meaningful components that shed new light on the Pi 2 pulsations of the geomagnetic field.
    Neural networks: the official journal of the International Neural Network Society 04/2012; 33:7-20. · 1.88 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We consider to learn a causal ordering of variables in a linear non-Gaussian acyclic model called LiNGAM. Several existing methods have been shown to consistently estimate a causal ordering assuming that all the model assumptions are correct. But, the estimation results could be distorted if some assumptions actually are violated. In this paper, we propose a new algorithm for learning causal orders that is robust against one typical violation of the model assumptions: latent confounders. We demonstrate the effectiveness of our method using artificial data.
    04/2012;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Discovering causal relations among observed variables in a given data set is a main topic in studies of statistics and artificial intelligence. Recently, some techniques to discover an identifiable causal structure have been explored based on non-Gaussianity of the observed data distribution. However, most of these are limited to continuous data. In this paper, we present a novel causal model for binary data and propose a new approach to derive an identifiable causal structure governing the data based on skew Bernoulli distributions of external noise. Experimental evaluation shows excellent performance for both artificial and real world data sets.
    02/2012;
  • Chris Clifton, Takashi Washio
    [Show abstract] [Hide abstract]
    ABSTRACT: Matrix factorization is a very powerful tool to find graph patterns, e.g. communities, anomalies, etc. A recent trend is to improve the usability of the discovered graph patterns, by encoding some interpretation-friendly properties (e.g., non-negativity, ...
    Statistical Analysis and Data Mining 02/2012; 5(1).
  • Source
    Akihiro Inokuchi, Hiroaki Ikuta, Takashi Washio
    [Show abstract] [Hide abstract]
    ABSTRACT: The mining of frequent subgraphs from labeled graph data has been studied extensively. Furthermore, much attention has recently been paid to frequent pattern mining from graph sequences. A method, called GTRACE, has been proposed to mine frequent patterns from graph sequences under the assumption that changes in graphs are gradual. Although GTRACE mines the frequent patterns efficiently, it still needs substantial computation time to mine the patterns from graph sequences containing large graphs and long sequences. In this paper, we propose a new version of GTRACE that enables efficient mining of frequent patterns based on the principle of a reverse search. The underlying concept of the reverse search is a general scheme for designing efficient algorithms for hard enumeration problems. Our performance study shows that the proposed method is efficient and scalable for mining both long and large graph sequence patterns and is several orders of magnitude faster than the original GTRACE.
    IEICE Transactions on Information and Systems. 10/2011; E95.D(7).

Publication Stats

1k Citations
38.50 Total Impact Points

Institutions

  • 1970–2014
    • Osaka University
      • • Institute of Scientific and Industrial Research
      • • The Institute of Scientific and Industrial Research (ISIR)
      Suika, Ōsaka, Japan
  • 2004–2012
    • National Cerebral and Cardiovascular Center
      • Department of Cardiovascular Medicine
      Ōsaka-shi, Osaka-fu, Japan
  • 2007
    • University of Grenoble
      Grenoble, Rhône-Alpes, France
  • 2006
    • Washington State University
      Pullman, Washington, United States
  • 2005
    • Hokkaido University
      Sapporo, Hokkaidō, Japan
  • 2001–2003
    • Kwansei Gakuin University
      Nishinomiya, Hyōgo, Japan
  • 1985–1993
    • Tohoku University
      Japan