English

Downward self-reducibility in the total function polynomial hierarchy

Computational Complexity 2025-07-28 v1 Data Structures and Algorithms

Abstract

A problem P\mathcal{P} is considered downward self-reducible, if there exists an efficient algorithm for P\mathcal{P} that is allowed to make queries to only strictly smaller instances of P\mathcal{P}. Downward self-reducibility has been well studied in the case of decision problems, and it is well known that any downward self-reducible problem must lie in PSPACE\mathsf{PSPACE}. Harsha, Mitropolsky and Rosen [ITCS, 2023] initiated the study of downward self reductions in the case of search problems. They showed the following interesting collapse: if a problem is in TFNP\mathsf{TFNP} and also downward self-reducible, then it must be in PLS\mathsf{PLS}. Moreover, if the problem admits a unique solution then it must be in UEOPL\mathsf{UEOPL}. We demonstrate that this represents just the tip of a much more general phenomenon, which holds for even harder search problems that lie higher up in the total function polynomial hierarchy (TFΣiP\mathsf{TF\Sigma_i^P}). In fact, even if we allow our downward self-reduction to be much more powerful, such a collapse will still occur. We show that any problem in TFΣiP\mathsf{TF\Sigma_i^P} which admits a randomized downward self-reduction with access to a Σi1P\mathsf{\Sigma_{i-1}^P} oracle must be in PLSΣi1P\mathsf{PLS}^{\mathsf{\Sigma_{i-1}^P}}. If the problem has \textit{essentially unique solutions} then it lies in UEOPLΣi1P\mathsf{UEOPL}^{\mathsf{\Sigma_{i-1}^P}}. As one (out of many) application of our framework, we get new upper bounds for the problems RangeAvoidance\mathrm{Range Avoidance} and LinearOrderingPrinciple\mathrm{Linear Ordering Principle} and show that they are both in UEOPLNP\mathsf{UEOPL}^{\mathsf{NP}}.

Keywords

Cite

@article{arxiv.2507.19108,
  title  = {Downward self-reducibility in the total function polynomial hierarchy},
  author = {Karthik Gajulapalli and Surendra Ghentiyala and Zeyong Li and Sidhant Saraogi},
  journal= {arXiv preprint arXiv:2507.19108},
  year   = {2025}
}
R2 v1 2026-07-01T04:18:33.983Z