Reduction of metric-sensitive logarithmic factors

Determine whether the $\log^2 n$ factors in the metric-sensitive competitive guarantee for Robustified Greedy can be reduced.

Background

Robustified Greedy achieves a metric-sensitive competitive ratio of O(k1−1/dlog⁡2n)O(k^{1-1/d}\log^2 n) in fixed-dimensional Euclidean spaces, and more generally an input-sensitive guarantee involving a log⁡2n\log^2 n factor. The paper establishes this bound but does not determine whether the logarithmic dependence is necessary or can be improved.

The authors identify reduction of these logarithmic factors as a direction for future work, making the unresolved issue explicit.

References

Natural directions for future work include narrowing the remaining gap to the $(2k-1)$ lower bound and determining whether the metric-sensitive logarithmic factors can be reduced.

— A Robustified Greedy Algorithm for Online Transportation with Improved Competitive Guarantees  (2609.39052 - Seth et al., 30 Sep 2026) in Conclusion