arXiv subjects

Joe Suzuki

Publications and source records attributed to Joe Suzuki.

At least 19 recordsLinked to original sources

arXiv2026

Bayesian ICA for Causal Discovery

LiNGAM identifies causal orders by exploiting the non-Gaussianity and mutual independence of structural disturbances. Several extensions allow particular forms of latent confounding, but a general criterion for comparing complete causal orders when the disturbances are dependent remains lacking. We formulate ICA in a Bayesian manner and propose Bayesian LiNGAM. For each candidate order, we use total correlation, or multivariate mutual information, among the order-dependent disturbances as a quantitative measure of confounding and estimate it from data by Bayesian marginal likelihoods. The causal order is selected by globally minimizing this estimate. We prove the consistency of the estimator and implement the global optimization using shortest-path search and dynamic programming. In the no-confounding large-sample regime, the search asymptotically requires the same number of score evaluations as Direct-LiNGAM. Bayesian LiNGAM therefore directly and globally optimizes the original ICA dependence criterion, rather than relying on greedy local decisions or a negentropy-based surrogate. Numerical experiments show that it generally achieves better causal-order recovery, particularly when latent confounding is present.

cs.IT↗

arXiv2025

Spacing Test for Fused Lasso

Detecting changepoints in a one-dimensional signal is a classical yet fundamental problem. The fused lasso provides an elegant convex formulation that produces a stepwise estimate of the mean, but quantifying the uncertainty of the detected changepoints remains difficult. Post-selection inference (PSI) offers a principled way to compute valid $p$-values after a data-driven selection, but its application to the fused lasso has been considered computationally cumbersome, requiring the tracking of many ``hit'' and ``leave'' events along the regularization path. In this paper, we show that the one-dimensional fused lasso has a surprisingly simple geometry: each changepoint enters in a strictly one-sided fashion, and there are no leave events. This structure implies that the so-called \emph{conservative spacing test} of Tibshirani et al.\ (2016), previously regarded as an approximation, is in fact \emph{exact}. The truncation region in the selective law reduces to a single lower bound given by the next knot on the LARS path. As a result, the exact selective $p$-value takes a closed form identical to the simple spacing statistic used in the LARS/lasso setting, with no additional computation. This finding establishes one of the rare cases in which an exact PSI procedure for the generalized lasso admits a closed-form pivot. We further validate the result by simulations and real data, confirming both exact calibration and high power. Keywords: fused lasso; changepoint detection; post-selection inference; spacing test; monotone LASSO

math.ST↗

arXiv2025

Estimation of the Learning Coefficient Using Empirical Loss

The learning coefficient plays a crucial role in analyzing the performance of information criteria, such as the Widely Applicable Information Criterion (WAIC) and the Widely Applicable Bayesian Information Criterion (WBIC), which Sumio Watanabe developed to assess model generalization ability. In regular statistical models, the learning coefficient is given by d/2, where d is the dimension of the parameter space. More generally, it is defined as the absolute value of the pole order of a zeta function derived from the Kullback-Leibler divergence and the prior distribution. However, except for specific cases such as reduced-rank regression, the learning coefficient cannot be derived in a closed form. Watanabe proposed a numerical method to estimate the learning coefficient, which Imai further refined to enhance its convergence properties. These methods utilize the asymptotic behavior of WBIC and have been shown to be statistically consistent as the sample size grows. In this paper, we propose a novel numerical estimation method that fundamentally differs from previous approaches and leverages a new quantity, "Empirical Loss," which was introduced by Watanabe. Through numerical experiments, we demonstrate that our proposed method exhibits both lower bias and lower variance compared to those of Watanabe and Imai. Additionally, we provide a theoretical analysis that elucidates why our method outperforms existing techniques and present empirical evidence that supports our findings.

stat.ML↗

arXiv2024

An Efficient Procedure for Computing Bayesian Network Structure Learning

We propose a globally optimal Bayesian network structure discovery algorithm based on a progressively leveled scoring approach. Bayesian network structure discovery is a fundamental yet NP-hard problem in the field of probabilistic graphical models, and as the number of variables increases, memory usage grows exponentially. The simple and effective method proposed by Silander and Myllym\"aki has been widely applied in this field, as it incrementally calculates local scores to achieve global optimality. However, existing methods that utilize disk storage, while capable of handling networks with a larger number of variables, introduce issues such as latency, fragmentation, and additional overhead associated with disk I/O operations. To avoid these problems, we explore how to further enhance computational efficiency and reduce peak memory usage using only memory. We introduce an efficient hierarchical computation method that requires only a single traversal of all local structures, retaining only the data and information necessary for the current computation, thereby improving efficiency and significantly reducing memory requirements. Experimental results indicate that our method, when using only memory, not only reduces peak memory usage but also improves computational efficiency compared to existing methods, demonstrating good scalability for handling larger networks and exhibiting stable experimental results. Ultimately, we successfully achieved the processing of a Bayesian network with 28 variables using only memory.

stat.ML↗

arXiv2024

Learning under Singularity: An Information Criterion improving WBIC and sBIC

We introduce a novel Information Criterion (IC), termed Learning under Singularity (LS), designed to enhance the functionality of the Widely Applicable Bayes Information Criterion (WBIC) and the Singular Bayesian Information Criterion (sBIC). LS is effective without regularity constraints and demonstrates stability. Watanabe defined a statistical model or a learning machine as regular if the mapping from a parameter to a probability distribution is one-to-one and its Fisher information matrix is positive definite. In contrast, models not meeting these conditions are termed singular. Over the past decade, several information criteria for singular cases have been proposed, including WBIC and sBIC. WBIC is applicable in non-regular scenarios but faces challenges with large sample sizes and redundant estimation of known learning coefficients. Conversely, sBIC is limited in its broader application due to its dependence on maximum likelihood estimates. LS addresses these limitations by enhancing the utility of both WBIC and sBIC. It incorporates the empirical loss from the Widely Applicable Information Criterion (WAIC) to represent the goodness of fit to the statistical model, along with a penalty term similar to that of sBIC. This approach offers a flexible and robust method for model selection, free from regularity constraints.

stat.ML↗

arXiv2024

Generalization of LiNGAM that allows confounding

LiNGAM determines the variable order from cause to effect using additive noise models, but it faces challenges with confounding. Previous methods maintained LiNGAM's fundamental structure while trying to identify and address variables affected by confounding. As a result, these methods required significant computational resources regardless of the presence of confounding, and they did not ensure the detection of all confounding types. In contrast, this paper enhances LiNGAM by introducing LiNGAM-MMI, a method that quantifies the magnitude of confounding using KL divergence and arranges the variables to minimize its impact. This method efficiently achieves a globally optimal variable order through the shortest path problem formulation. LiNGAM-MMI processes data as efficiently as traditional LiNGAM in scenarios without confounding while effectively addressing confounding situations. Our experimental results suggest that LiNGAM-MMI more accurately determines the correct variable order, both in the presence and absence of confounding. The code is in the supplementary file in this link: https://github.com/SkyJoyTianle/ISIT2024.

cs.LG↗

arXiv2024

Functional Linear Non-Gaussian Acyclic Model for Causal Discovery

In causal discovery, non-Gaussianity has been used to characterize the complete configuration of a Linear Non-Gaussian Acyclic Model (LiNGAM), encompassing both the causal ordering of variables and their respective connection strengths. However, LiNGAM can only deal with the finite-dimensional case. To expand this concept, we extend the notion of variables to encompass vectors and even functions, leading to the Functional Linear Non-Gaussian Acyclic Model (Func-LiNGAM). Our motivation stems from the desire to identify causal relationships in brain-effective connectivity tasks involving, for example, fMRI and EEG datasets. We demonstrate why the original LiNGAM fails to handle these inherently infinite-dimensional datasets and explain the availability of functional data analysis from both empirical and theoretical perspectives. {We establish theoretical guarantees of the identifiability of the causal relationship among non-Gaussian random vectors and even random functions in infinite-dimensional Hilbert spaces.} To address the issue of sparsity in discrete time points within intrinsic infinite-dimensional functional data, we propose optimizing the coordinates of the vectors using functional principal component analysis. Experimental results on synthetic data verify the ability of the proposed framework to identify causal relationships among multivariate functions using the observed samples. For real data, we focus on analyzing the brain connectivity patterns derived from fMRI data.

cs.LG↗

arXiv2023

Post-Selection Inference for Sparse Estimation

When the model is not known and parameter testing or interval estimation is conducted after model selection, it is necessary to consider selective inference. This paper discusses this issue in the context of sparse estimation. Firstly, we describe selective inference related to Lasso as per \cite{lee}, and then present polyhedra and truncated distributions when applying it to methods such as Forward Stepwise and LARS. Lastly, we discuss the Significance Test for Lasso by \cite{significant} and the Spacing Test for LARS by \cite{ryan_exact}. This paper serves as a review article. Keywords: post-selective inference, polyhedron, LARS, lasso, forward stepwise, significance test, spacing test.

stat.ME↗

arXiv2023

Newton-type Methods with the Proximal Gradient Step for Sparse Estimation

In this paper, we propose new methods to efficiently solve convex optimization problems encountered in sparse estimation, which include a new quasi-Newton method that avoids computing the Hessian matrix and improves efficiency, and we prove its fast convergence. We also prove the local convergence of the Newton method under weaker assumptions. Our proposed methods offer a more efficient and effective approach, particularly for L1 regularization and group regularization problems, as they involve variable selection with each update. Through numerical experiments, we demonstrate the efficiency of our methods in solving problems encountered in sparse estimation. Our contributions include theoretical guarantees and practical applications for various problems.

math.OC↗

arXiv2023

Dropout Drops Double Descent

This study demonstrates that double descent can be mitigated by adding a dropout layer adjacent to the fully connected linear layer. The unexpected double-descent phenomenon garnered substantial attention in recent years, resulting in fluctuating prediction error rates as either sample size or model size increases. Our paper posits that the optimal test error, in terms of the dropout rate, shows a monotonic decrease in linear regression with increasing sample size. Although we do not provide a precise mathematical proof of this statement, we empirically validate through experiments that the test error decreases for each dropout rate. The statement we prove is that the expected test error for each dropout rate within a certain range decreases when the dropout rate is fixed. Our experimental results substantiate our claim, showing that dropout with an optimal dropout rate can yield a monotonic test error curve in nonlinear neural networks. These experiments were conducted using the Fashion-MNIST and CIFAR-10 datasets. These findings imply the potential benefit of incorporating dropout into risk curve scaling to address the peak phenomenon. To our knowledge, this study represents the first investigation into the relationship between dropout and double descent.

cs.LG↗

arXiv2022

Converting ADMM to a Proximal Gradient for Efficient Sparse Estimation

In sparse estimation, such as fused lasso and convex clustering, we apply either the proximal gradient method or the alternating direction method of multipliers (ADMM) to solve the problem. It takes time to include matrix division in the former case, while an efficient method such as FISTA (fast iterative shrinkage-thresholding algorithm) has been developed in the latter case. This paper proposes a general method for converting the ADMM solution to the proximal gradient method, assuming that assumption that the derivative of the objective function is Lipschitz continuous. Then, we apply it to sparse estimation problems, such as sparse convex clustering and trend filtering, and we show by numerical experiments that we can obtain a significant improvement in terms of efficiency.

math.OC↗

arXiv2021

Causal Order Identification to Address Confounding: Binary Variables

This paper considers an extension of the linear non-Gaussian acyclic model (LiNGAM) that determines the causal order among variables from a dataset when the variables are expressed by a set of linear equations, including noise. In particular, we assume that the variables are binary. The existing LiNGAM assumes that no confounding is present, which is restrictive in practice. Based on the concept of independent component analysis (ICA), this paper proposes an extended framework in which the mutual information among the noises is minimized. Another significant contribution is to reduce the realization of the shortest path problem. The distance between each pair of nodes expresses an associated mutual information value, and the path with the minimum sum (KL divergence) is sought. Although $p!$ mutual information values should be compared, this paper dramatically reduces the computation when no confounding is present. The proposed algorithm finds the globally optimal solution, while the existing locally greedily seek the order based on hypothesis testing. We use the best estimator in the sense of Bayes/MDL that correctly detects independence for mutual information estimation. Experiments using artificial and actual data show that the proposed version of LiNGAM achieves significantly better performance, particularly when confounding is present.

cs.LG↗

arXiv2021

Efficient proximal gradient algorithms for joint graphical lasso

We consider learning an undirected graphical model from sparse data. While several efficient algorithms have been proposed for graphical lasso (GL), the alternating direction method of multipliers (ADMM) is the main approach taken concerning for joint graphical lasso (JGL). We propose proximal gradient procedures with and without a backtracking option for the JGL. These procedures are first-order and relatively simple, and the subproblems are solved efficiently in closed form. We further show the boundedness for the solution of the JGL problem and the iterations in the algorithms. The numerical results indicate that the proposed algorithms can achieve high accuracy and precision, and their efficiency is competitive with state-of-the-art algorithms.

math.OC↗

arXiv2018

Forest Learning from Data and its Universal Coding

This paper considers structure learning from data with $n$ samples of $p$ variables, assuming that the structure is a forest, using the Chow-Liu algorithm. Specifically, for incomplete data, we construct two model selection algorithms that complete in $O(p^2)$ steps: one obtains a forest with the maximum posterior probability given the data, and the other obtains a forest that converges to the true one as $n$ increases. We show that the two forests are generally different when some values are missing. Additionally, we present estimations for benchmark data sets to demonstrate that both algorithms work in realistic situations. Moreover, we derive the conditional entropy provided that no value is missing, and we evaluate the per-sample expected redundancy for the universal coding of incomplete data in terms of the number of non-missing samples.

cs.IT↗

arXiv2017

Klein's Fundamental 2-Form of Second Kind for the $C_{ab}$ Curves

In this paper, we derive the exact formula of Klein's fundamental 2-form of second kind for the so-called $C_{ab}$ curves. The problem was initially solved by Klein in the 19th century for the hyper-elliptic curves, but little progress had been seen for its extension for more than 100 years. Recently, it has been addressed by several authors, and was solved for subclasses of the $C_{ab}$ curves whereas they found a way to find its individual solution numerically. The formula gives a standard cohomological basis for the curves, and has many applications in algebraic geometry, physics, and applied mathematics, not just analyzing sigma functions in a general way.

math.AG↗

arXiv2016

A Theoretical Analysis of the BDeu Scores in Bayesian Network Structure Learning

In Bayesian network structure learning (BNSL), we need the prior probability over structures and parameters. If the former is the uniform distribution, the latter determines the correctness of BNSL. In this paper, we compare BDeu (Bayesian Dirichlet equivalent uniform) and Jeffreys' prior w.r.t. their consistency. When we seek a parent set $U$ of a variable $X$, we require regularity that if $H(X|U)\leq H(X|U')$ and $U\subsetneq U'$, then $U$ should be chosen rather than $U'$. We prove that the BDeu scores violate the property and cause fatal situations in BNSL. This is because for the BDeu scores, for any sample size $n$,there exists a probability in the form $P(X,Y,Z)={P(XZ)P(YZ)}/{P(Z)}$ such that the probability of deciding that $X$ and $Y$ are not conditionally independent given $Z$ is more than a half. For Jeffreys' prior, the false-positive probability uniformly converges to zero without depending on any parameter values, and no such an inconvenience occurs.

cs.LG↗

arXiv2015

Miura: Divisor Class Group Arithmetic

The Package Miura contains functions that compute divisor class group arithmetic for nonsingular curves. The package reduces computation in a divisor class group to that in the ideal class group via the isomorphism. The underlying quotient ring should be over the ideal given by a nonsingular curve in the form of Miura. Although computing the multiplication of two integral ideals is not hard, we need to obtain an ideal such that the shortest Groebner basis component is the minimum in order to obtain the representative of the ideal class. Although the basic procedure is due to Arita, the source code has become much shorter using MaCaulay2. The package is useful not just for computation itself but also for understanding the divisor class group arithmetic from the ideal point of view.

math.AC↗

arXiv2014

Universal Bayesian Measures and Universal Histogram Sequences

Consider universal data compression: the length $l(x^n)$ of sequence $x^n\in A^n$ with finite alphabet $A$ and length $n$ satisfies Kraft's inequality over $A^n$, and $-\frac{1}{n}\log \frac{P^n(x^n)}{Q^n(x^n)}$ almost surely converges to zero as $n$ grows for the $Q^n(x^n)=2^{-l(x^n)}$ and any stationary ergodic source $P$. In this paper, we say such a $Q$ is a universal Bayesian measure. We generalize the notion to the sources in which the random variables may be either discrete, continuous, or none of them. The basic idea is due to Boris Ryabko who utilized model weighting over histograms that approximate $P$, assuming that a density function of $P$ exists. However, the range of $P$ depends on the choice of the histogram sequence. The universal Bayesian measure constructed in this paper overcomes the drawbacks and has many applications to infer relation among random variables, and extends the application area of the minimum description length principle.

cs.IT↗