Match forecasts in UEFA club competitions: Elo ratings versus Transfermarkt valuations
该研究通过比较Elo评分和转会市场估值在预测2020/21至2024/25赛季欧洲足球俱乐部比赛结果中的表现，发现两者单独使用时准确率相近，结合使用略有提升。
0 citationsRead paper
Selected work
Representative Papers
Individual Rationality in Constrained Hedonic Games: Friends, Enemies, and Neutrals
This study investigates the computational complexity of individually rational coalition structures in hedonic games under size constraints. Through classical and parameterized complexity analyses combined with graph coloring reductions, it reveals that adversarial structures fundamentally drive intractability in friend-oriented models. The research comprehensively characterizes the boundary conditions under which symmetry and size constraints influence complexity, establishing a systematic taxonomy of computational hardness. It identifies the critical structural properties responsible for NP-hardness while delineating tractable cases solvable in polynomial time. These findings provide a rigorous theoretical foundation for algorithm design in constrained cooperative games, clarifying precisely when efficient computation is feasible versus inherently intractable within this important class of hedonic game models.
0 citationsRead paper
Workplace dependence in urban economies
The widespread adoption of remote work has exacerbated intra-urban inequalities in health risks, social interaction, and economic opportunity. Leveraging high-resolution hourly human mobility data and firm registration records, this study exploits variations in pandemic-related mobility restrictions as a quasi-natural experiment to investigate the drivers of workplace dependence through multidimensional regression and spatial heterogeneity models. The analysis reveals that industry type and firm productivity are key determinants. Moreover, income and gender effects are significantly moderated by distance from the city center, giving rise to a “service trap” in core urban areas: neighborhoods characterized by female-dominated employment and diverse income sources exhibit heightened reliance on in-person attendance, thereby extending remote-work disparities from the individual level to the broader urban ecosystem.
0 citationsRead paper
Necessary President in Elections with Parties
This study investigates the computational complexity of the “necessary winner” problem in partisan elections: determining whether a given candidate, once nominated by their party, wins under every possible combination of nominations by other parties. We provide the first systematic characterization of this problem’s complexity across a range of voting rules, employing both classical and parameterized complexity theory—including coNP-completeness and W[1]/W[2]-hardness results. Our main findings reveal that the problem remains computationally intractable even under highly restricted settings, such as when each party has only two candidates or the number of voters is fixed. In contrast, the problem is polynomial-time solvable under Borda, Maximin, and Copeland^α rules, yet becomes coNP-complete and parameterized intractable under positional scoring rules like ℓ-Approval and ℓ-Veto, as well as under the Ranked Pairs rule.
0 citationsRead paper
Impact of AI Tools on Learning Outcomes: Decreasing Knowledge and Over-Reliance
This study investigates the deep pedagogical impacts of generative AI tools on student learning. Motivated by concerns over diminished motivation, superficial knowledge acquisition, and cognitive substitution arising from students’ overreliance on AI for assignments and assessments in operations research courses, we conducted a randomized controlled trial: one group was permitted unrestricted AI use, while the other was prohibited from using AI throughout the course. Integrating grade compensation mechanisms, quantitative academic performance analysis, and educational-psychological behavioral observation, we provide the first empirical evidence—within a controlled instructional setting—that unfettered AI use significantly reduces classroom engagement, impairs conceptual mastery, and triggers systemic cognitive degradation. Students exhibit entrenched path dependence, undermining traditional assessment validity. Beyond establishing a causal link between AI misuse and declining learning quality, the study introduces the “cognitive substitution effect” as a novel theoretical framework, offering critical empirical foundations for rethinking educational interventions and assessment design in the age of artificial intelligence.
0 citationsRead paper
Recent publications
Latest Papers
Match forecasts in UEFA club competitions: Elo ratings versus Transfermarkt valuations
该研究通过比较Elo评分和转会市场估值在预测2020/21至2024/25赛季欧洲足球俱乐部比赛结果中的表现，发现两者单独使用时准确率相近，结合使用略有提升。
0 citationsRead paper
Individual Rationality in Constrained Hedonic Games: Friends, Enemies, and Neutrals
This study investigates the computational complexity of individually rational coalition structures in hedonic games under size constraints. Through classical and parameterized complexity analyses combined with graph coloring reductions, it reveals that adversarial structures fundamentally drive intractability in friend-oriented models. The research comprehensively characterizes the boundary conditions under which symmetry and size constraints influence complexity, establishing a systematic taxonomy of computational hardness. It identifies the critical structural properties responsible for NP-hardness while delineating tractable cases solvable in polynomial time. These findings provide a rigorous theoretical foundation for algorithm design in constrained cooperative games, clarifying precisely when efficient computation is feasible versus inherently intractable within this important class of hedonic game models.
0 citationsRead paper
Workplace dependence in urban economies
The widespread adoption of remote work has exacerbated intra-urban inequalities in health risks, social interaction, and economic opportunity. Leveraging high-resolution hourly human mobility data and firm registration records, this study exploits variations in pandemic-related mobility restrictions as a quasi-natural experiment to investigate the drivers of workplace dependence through multidimensional regression and spatial heterogeneity models. The analysis reveals that industry type and firm productivity are key determinants. Moreover, income and gender effects are significantly moderated by distance from the city center, giving rise to a “service trap” in core urban areas: neighborhoods characterized by female-dominated employment and diverse income sources exhibit heightened reliance on in-person attendance, thereby extending remote-work disparities from the individual level to the broader urban ecosystem.
0 citationsRead paper
Necessary President in Elections with Parties
This study investigates the computational complexity of the “necessary winner” problem in partisan elections: determining whether a given candidate, once nominated by their party, wins under every possible combination of nominations by other parties. We provide the first systematic characterization of this problem’s complexity across a range of voting rules, employing both classical and parameterized complexity theory—including coNP-completeness and W[1]/W[2]-hardness results. Our main findings reveal that the problem remains computationally intractable even under highly restricted settings, such as when each party has only two candidates or the number of voters is fixed. In contrast, the problem is polynomial-time solvable under Borda, Maximin, and Copeland^α rules, yet becomes coNP-complete and parameterized intractable under positional scoring rules like ℓ-Approval and ℓ-Veto, as well as under the Ranked Pairs rule.
0 citationsRead paper
Impact of AI Tools on Learning Outcomes: Decreasing Knowledge and Over-Reliance
This study investigates the deep pedagogical impacts of generative AI tools on student learning. Motivated by concerns over diminished motivation, superficial knowledge acquisition, and cognitive substitution arising from students’ overreliance on AI for assignments and assessments in operations research courses, we conducted a randomized controlled trial: one group was permitted unrestricted AI use, while the other was prohibited from using AI throughout the course. Integrating grade compensation mechanisms, quantitative academic performance analysis, and educational-psychological behavioral observation, we provide the first empirical evidence—within a controlled instructional setting—that unfettered AI use significantly reduces classroom engagement, impairs conceptual mastery, and triggers systemic cognitive degradation. Students exhibit entrenched path dependence, undermining traditional assessment validity. Beyond establishing a causal link between AI misuse and declining learning quality, the study introduces the “cognitive substitution effect” as a novel theoretical framework, offering critical empirical foundations for rethinking educational interventions and assessment design in the age of artificial intelligence.
0 citationsRead paper