Searcharxiv⌕ Search

arXiv subjects

Yuya Sasaki

Publications and source records attributed to Yuya Sasaki.

At least 37 records · Page 2Linked to original sources
arXiv2025

Benchmarking Fairness-aware Graph Neural Networks in Knowledge Graphs

Graph neural networks (GNNs) are powerful tools for learning from graph-structured data but often produce biased predictions with respect to sensitive attributes. Fairness-aware GNNs have been actively studied for mitigating biased predictions. However, no prior studies have evaluated fairness-aware GNNs on knowledge graphs, which are one of the most important graphs in many applications, such as recommender systems. Therefore, we introduce a benchmarking study on knowledge graphs. We generate new graphs from three knowledge graphs, YAGO, DBpedia, and Wikidata, that are significantly larger than the existing graph datasets used in fairness studies. We benchmark inprocessing and preprocessing methods in different GNN backbones and early stopping conditions. We find several key insights: (i) knowledge graphs show different trends from existing datasets; clearer trade-offs between prediction accuracy and fairness metrics than other graphs in fairness-aware GNNs, (ii) the performance is largely affected by not only fairness-aware GNN methods but also GNN backbones and early stopping conditions, and (iii) preprocessing methods often improve fairness metrics, while inprocessing methods improve prediction accuracy.

cs.LG↗
arXiv2025

Genuinely Robust Inference for Clustered Data

Conventional cluster-robust inference can be invalid when data contain clusters of unignorably large size. We formalize this issue by deriving a necessary and sufficient condition for its validity, and show that this condition is frequently violated in practice: specifications from 77% of empirical research articles in American Economic Review and Econometrica during 2020-2021 appear not to meet it. To address this limitation, we propose a genuinely robust inference procedure based on a new cluster score bootstrap. We establish its validity and size control across broad classes of data-generating processes where conventional methods break down. Simulation studies corroborate our theoretical findings, and empirical applications illustrate that employing the proposed method can substantially alter conventional statistical conclusions.

econ.EM↗
arXiv2025

Algorithmic Fairness: Not a Purely Technical but Socio-Technical Property

The rapid trend of deploying artificial intelligence (AI) and machine learning (ML) systems in socially consequential domains has raised growing concerns about their trustworthiness, including potential discriminatory behaviours. Research in algorithmic fairness has generated a proliferation of mathematical definitions and metrics, yet persistent misconceptions and limitations -- both within and beyond the fairness community -- limit their effectiveness, such as an unreached consensus on its understanding, prevailing measures primarily tailored to binary group settings, and superficial handling for intersectional contexts. Here we critically remark on these misconceptions and argue that fairness cannot be reduced to purely technical constraints on models; we also examine the limitations of existing fairness measures through conceptual analysis and empirical illustrations, showing their limited applicability in the face of complex real-world scenarios, challenging prevailing views on the incompatibility between accuracy and fairness as well as that among fairness measures themselves, and outlining three worth-considering principles in the design of fairness measures. We believe these findings will help bridge the gap between technical formalisation and social realities and meet the challenges of real-world AI/ML deployment.

cs.LG↗
arXiv2025

On real forms of Fermat hypersurfaces

In this paper, we compute the number of real forms of Fermat hypersurfaces for degree $d \ge 2$ except the degree 4 surface case, and give explicit descriptions of them.

math.AG↗
arXiv2025

Public Perceptions of Fairness Metrics Across Borders

Which fairness metrics are appropriately applicable in your contexts? There may be instances of discordance regarding the perception of fairness, even when the outcomes comply with established fairness metrics. Several questionnaire-based surveys have been conducted to evaluate fairness metrics with human perceptions of fairness. However, these surveys were limited in scope, including only a few hundred participants within a single country. In this study, we conduct an international survey to evaluate public perceptions of various fairness metrics in decision-making scenarios. We collected responses from 1,000 participants in each of China, France, Japan, and the United States, amassing a total of 4,000 participants, to analyze the preferences of fairness metrics. Our survey consists of three distinct scenarios paired with four fairness metrics. This investigation explores the relationship between personal attributes and the choice of fairness metrics, uncovering a significant influence of national context on these preferences.

cs.AI↗
arXiv2025

Matching $\leq$ Hybrid $\leq$ Difference in Differences

Since LaLonde's (1986) seminal paper, there has been ongoing interest in estimating treatment effects using pre- and post-intervention data. Scholars have traditionally used experimental benchmarks to evaluate the accuracy of alternative econometric methods, including Matching, Difference-in-Differences (DID), and their hybrid forms (e.g., Heckman et al., 1998b; Dehejia and Wahba, 2002; Smith and Todd, 2005). We revisit these methodologies in the evaluation of job training and educational programs using four datasets (LaLonde, 1986; Heckman et al., 1998a; Smith and Todd, 2005; Chetty et al., 2014a; Athey et al., 2020), and show that the inequality relationship, Matching $\leq$ Hybrid $\leq$ DID, appears as a consistent norm, rather than a mere coincidence. We provide a formal theoretical justification for this puzzling phenomenon under plausible conditions such as negative selection, by generalizing the classical bracketing (Angrist and Pischke, 2009, Section 5). Consequently, when treatments are expected to be non-negative, DID tends to provide optimistic estimates, while Matching offers more conservative ones.

econ.EM↗
arXiv2025

Non-Robustness of the Cluster-Robust Inference: with a Proposal of a New Robust Method

The conventional cluster-robust (CR) standard errors may not be robust. They are vulnerable to data that contain a small number of large clusters. When a researcher uses the 51 states in the U.S. as clusters, the largest cluster (California) consists of about 10% of the total sample. Such a case in fact violates the assumptions under which the widely used CR methods are guaranteed to work. We formally show that the conventional CR methods fail if the distribution of cluster sizes follows a power law with exponent less than two. Besides the example of 51 state clusters, some examples are drawn from a list of recent original research articles published in a top journal. In light of these negative results about the existing CR methods, we propose a weighted CR (WCR) method as a simple fix. Simulation studies support our arguments that the WCR method is robust while the conventional CR methods are not.

econ.EM↗
arXiv2024

Mining Path Association Rules in Large Property Graphs (with Appendix)

How can we mine frequent path regularities from a graph with edge labels and vertex attributes? The task of association rule mining successfully discovers regular patterns in item sets and substructures. Still, to our best knowledge, this concept has not yet been extended to path patterns in large property graphs. In this paper, we introduce the problem of path association rule mining (PARM). Applied to any \emph{reachability path} between two vertices within a large graph, PARM discovers regular ways in which path patterns, identified by vertex attributes and edge labels, co-occur with each other. We develop an efficient and scalable algorithm PIONEER that exploits an anti-monotonicity property to effectively prune the search space. Further, we devise approximation techniques and employ parallelization to achieve scalable path association rule mining. Our experimental study using real-world graph data verifies the significance of path association rules and the efficiency of our solutions.

cs.DB↗
arXiv2024

Estimation and Inference for Causal Functions with Multiway Clustered Data

This paper proposes methods of estimation and uniform inference for a general class of causal functions, such as the conditional average treatment effects and the continuous treatment effects, under multiway clustering. The causal function is identified as a conditional expectation of an adjusted (Neyman-orthogonal) signal that depends on high-dimensional nuisance parameters. We propose a two-step procedure where the first step uses machine learning to estimate the high-dimensional nuisance parameters. The second step projects the estimated Neyman-orthogonal signal onto a dictionary of basis functions whose dimension grows with the sample size. For this two-step procedure, we propose both the full-sample and the multiway cross-fitting estimation approaches. A functional limit theory is derived for these estimators. To construct the uniform confidence bands, we develop a novel resampling procedure, called the multiway cluster-robust sieve score bootstrap, that extends the sieve score bootstrap (Chen and Christensen, 2018) to the novel setting with multiway clustering. Extensive numerical simulations showcase that our methods achieve desirable finite-sample behaviors. We apply the proposed methods to analyze the causal relationship between mistrust levels in Africa and the historical slave trade. Our analysis rejects the null hypothesis of uniformly zero effects and reveals heterogeneous treatment effects, with significant impacts at higher levels of trade volumes.

econ.EM↗
arXiv2024

Extreme Quantile Treatment Effects under Endogeneity: Evaluating Policy Effects for the Most Vulnerable Individuals

We introduce a novel method for estimating and conducting inference about extreme quantile treatment effects (QTEs) in the presence of endogeneity. Our approach is applicable to a broad range of empirical research designs, including instrumental variables design and regression discontinuity design, among others. By leveraging regular variation and subsampling, the method ensures robust performance even in extreme tails, where data may be sparse or entirely absent. Simulation studies confirm the theoretical robustness of our approach. Applying our method to assess the impact of job training provided by the Job Training Partnership Act (JTPA), we find significantly negative QTEs for the lowest quantiles (i.e., the most disadvantaged individuals), contrasting with previous literature that emphasizes positive QTEs for intermediate quantiles.

econ.EM↗
arXiv2024

Quantifying gendered citation imbalance in computer science conferences

The number of citations received by papers often exhibits imbalances in terms of author attributes such as country of affiliation and gender. While recent studies have quantified citation imbalance in terms of the authors' gender in journal papers, the computer science discipline, where researchers frequently present their work at conferences, may exhibit unique patterns in gendered citation imbalance. Additionally, understanding how network properties in citations influence citation imbalances remains challenging due to a lack of suitable reference models. In this paper, we develop a family of reference models for citation networks and investigate gender imbalance in citations between papers published in computer science conferences. By deploying these reference models, we found that homophily in citations is strongly associated with gendered citation imbalance in computer science, whereas heterogeneity in the number of citations received per paper has a relatively minor association with it. Furthermore, we found that the gendered citation imbalance is most pronounced in papers published in the highest-ranked conferences, is present across different subfields, and extends to citation-based rankings of papers. Our study provides a framework for investigating associations between network properties and citation imbalances, aiming to enhance our understanding of the structure and dynamics of citations between research publications.

cs.SI↗
arXiv2024

The Informativeness of Combined Experimental and Observational Data under Dynamic Selection

This paper addresses the challenge of estimating the Average Treatment Effect on the Treated Survivors (ATETS; Vikstrom et al., 2018) in the absence of long-term experimental data, utilizing available long-term observational data instead. We establish two theoretical results. First, it is impossible to obtain informative bounds for the ATETS with no model restriction and no auxiliary data. Second, to overturn this negative result, we explore as a promising avenue the recent econometric developments in combining experimental and observational data (e.g., Athey et al., 2020, 2019); we indeed find that exploiting short-term experimental data can be informative without imposing classical model restrictions. Furthermore, building on Chesher and Rosen (2017), we explore how to systematically derive sharp identification bounds, exploiting both the novel data-combination principles and classical model restrictions. Applying the proposed method, we explore what can be learned about the long-run effects of job training programs on employment without long-term experimental data.

econ.EM↗
arXiv2024

Non-Existent Moments of Earnings Growth

The literature often employs moment-based earnings risk measures like variance, skewness, and kurtosis. However, under heavy-tailed distributions, these moments may not exist in the population. Our empirical analysis reveals that population kurtosis, skewness, and variance often do not exist for the conditional distribution of earnings growth. This challenges moment-based analyses. We propose robust conditional Pareto exponents as novel earnings risk measures, developing estimation and inference methods. Using the UK New Earnings Survey Panel Dataset (NESPD) and US Panel Study of Income Dynamics (PSID), we find: 1) Moments often fail to exist; 2) Earnings risk increases over the life cycle; 3) Job stayers face higher earnings risk; 4) These patterns persist during the 2007--2008 recession and the 2015--2016 positive growth period.

econ.EM↗
arXiv2024

A Bracketing Relationship for Long-Term Policy Evaluation with Combined Experimental and Observational Data

Combining short-term experimental data with observational data enables credible long-term policy evaluation. The literature offers two key but non-nested assumptions, namely the latent unconfoundedness (LU; Athey et al., 2020) and equi-confounding bias (ECB; Ghassami et al., 2022) conditions, to correct observational selection. Committing to the wrong assumption leads to biased estimation. To mitigate such risks, we provide a novel bracketing relationship (cf. Angrist and Pischke, 2009) repurposed for the setting with data combination: the LU-based estimand and the ECB-based estimand serve as the lower and upper bounds, respectively, with the true causal effect lying in between if either assumption holds. For researchers further seeking point estimates, our Lalonde-style exercise suggests the conservatively more robust LU-based lower bounds align closely with the hold-out experimental estimates for educational policy evaluation. We investigate the economic substantives of these findings through the lens of a nonparametric class of selection mechanisms and sensitivity analysis. We uncover as key the sub-martingale property and sufficient-statistics role (Chetty, 2009) of the potential outcomes of student test scores (Chetty et al., 2011, 2014).

econ.EM↗
arXiv2023

Standard errors for two-way clustering with serially correlated time effects

We propose improved standard errors and an asymptotic distribution theory for two-way clustered panels. Our proposed estimator and theory allow for arbitrary serial dependence in the common time effects, which is excluded by existing two-way methods, including the popular two-way cluster standard errors of Cameron, Gelbach, and Miller (2011) and the cluster bootstrap of Menzel (2021). Our asymptotic distribution theory is the first which allows for this level of inter-dependence among the observations. Under weak regularity conditions, we demonstrate that the least squares estimator is asymptotically normal, our proposed variance estimator is consistent, and t-ratios are asymptotically standard normal, permitting conventional inference. We present simulation evidence that confidence intervals constructed with our proposed standard errors obtain superior coverage performance relative to existing methods. We illustrate the relevance of the proposed method in an empirical application to a standard Fama-French three-factor regression.

econ.EM↗
arXiv2023

A Simple and Scalable Graph Neural Network for Large Directed Graphs

Node classification is one of the hottest tasks in graph analysis. Though existing studies have explored various node representations in directed and undirected graphs, they have overlooked the distinctions of their capabilities to capture the information of graphs. To tackle the limitation, we investigate various combinations of node representations (aggregated features vs. adjacency lists) and edge direction awareness within an input graph (directed vs. undirected). We address the first empirical study to benchmark the performance of various GNNs that use either combination of node representations and edge direction awareness. Our experiments demonstrate that no single combination stably achieves state-of-the-art results across datasets, which indicates that we need to select appropriate combinations depending on the dataset characteristics. In response, we propose a simple yet holistic classification method A2DUG which leverages all combinations of node representations in directed and undirected graphs. We demonstrate that A2DUG stably performs well on various datasets and improves the accuracy up to 11.29 compared with the state-of-the-art methods. To spur the development of new methods, we publicly release our complete codebase under the MIT license.

cs.LG↗
arXiv2023

Learned spatial data partitioning

Due to the significant increase in the size of spatial data, it is essential to use distributed parallel processing systems to efficiently analyze spatial data. In this paper, we first study learned spatial data partitioning, which effectively assigns groups of big spatial data to computers based on locations of data by using machine learning techniques. We formalize spatial data partitioning in the context of reinforcement learning and develop a novel deep reinforcement learning algorithm. Our learning algorithm leverages features of spatial data partitioning and prunes ineffective learning processes to find optimal partitions efficiently. Our experimental study, which uses Apache Sedona and real-world spatial data, demonstrates that our method efficiently finds partitions for accelerating distance join queries and reduces the workload run time by up to 59.4%.

cs.DB↗