activity
20242026
collaborators

6 papers

math.OC2026

Accelerating operator Sinkhorn iteration with overrelaxation

Tasuku Soma, André Uschmajew

We propose accelerated versions of the operator Sinkhorn iteration for operator scaling using successive overrelaxation. We analyze the local convergence rates of these accelerated…

math.OC2026

Numerically stable variants of overrelaxation for operator Sinkhorn iteration

Henrik Eisenmann, Tasuku Soma, Xun Tang +1

We consider accelerated versions of the operator Sinkhorn iteration (OSI) for solving scaling problems for completely positive maps. Based on the interpretation of OSI as alternati…

cs.DS2025

-Approximation Algorithms for Bipartiteness Ratio

Tasuku Soma, Mingquan Ye, Yuichi Yoshida

We propose an -approximation algorithm for the bipartiteness ratio of undirected graphs introduced by Trevisan (SIAM Journal on Computing, vol. 41, no. 6, 2012), where $…

math.OC2025

Algorithmic aspects of semistability of quiver representations

Yuni Iwamasa, Taihei Oki, Tasuku Soma

We study the semistability of quiver representations from an algorithmic perspective. We present efficient algorithms for several fundamental computational problems on the semistab…

cs.LG2025

Near-Optimal Algorithms for Group Distributionally Robust Optimization and Beyond

Tasuku Soma, Khashayar Gatmiry, Sharut Gupta +1

Distributionally robust optimization (DRO) can improve the robustness and fairness of learning methods. In this paper, we devise stochastic algorithms for a class of DRO problems i…

math.OC2024

Algebraic Algorithms for Fractional Linear Matroid Parity via Non-commutative Rank

Taihei Oki, Tasuku Soma

Matrix representations are a powerful tool for designing efficient algorithms for combinatorial optimization problems such as matching, and linear matroid intersection and parity.…