English

Partition bound is quadratically tight for product distributions

Computational Complexity 2020-05-08 v4

Abstract

Let f:{0,1}n×{0,1}n{0,1}f : \{0,1\}^n \times \{0,1\}^n \rightarrow \{0,1\} be a 2-party function. For every product distribution μ\mu on {0,1}n×{0,1}n\{0,1\}^n \times \{0,1\}^n, we show that CC0.49μ(f)=O((logprt1/8(f)loglogprt1/8(f))2),\mathsf{CC}^\mu_{0.49}(f) = O\left(\left(\log \mathsf{prt}_{1/8}(f) \cdot \log \log \mathsf{prt}_{1/8}(f)\right)^2\right), where CCεμ(f)\mathsf{CC}^\mu_\varepsilon(f) is the distributional communication complexity of ff with error at most ε\varepsilon under the distribution μ\mu and prt1/8(f)\mathsf{prt}_{1/8}(f) is the {\em partition bound} of ff, as defined by Jain and Klauck [{\em Proc. 25th CCC}, 2010]. We also prove a similar bound in terms of IC1/8(f)\mathsf{IC}_{1/8}(f), the {\em information complexity} of ff, namely, CC0.49μ(f)=O((IC1/8(f)logIC1/8(f))2).\mathsf{CC}^\mu_{0.49}(f) = O\left(\left(\mathsf{IC}_{1/8}(f) \cdot \log \mathsf{IC}_{1/8}(f)\right)^2\right). The latter bound was recently and independently established by Kol [{\em Proc. 48th STOC}, 2016] using a different technique. We show a similar result for query complexity under product distributions. Let g:{0,1}n{0,1}g : \{0,1\}^n \rightarrow \{0,1\} be a function. For every bit-wise product distribution μ\mu on {0,1}n\{0,1\}^n, we show that QC0.49μ(g)=O((logqprt1/8(g)loglogqprt1/8(g))2),\mathsf{QC}^\mu_{0.49}(g) = O\left(\left( \log \mathsf{qprt}_{1/8}(g) \cdot \log \log\mathsf{qprt}_{1/8}(g) \right)^2 \right), where QCεμ(g)\mathsf{QC}^\mu_{\varepsilon}(g) is the distributional query complexity of ff with error at most ε\varepsilon under the distribution μ\mu and qprt1/8(g))\mathsf{qprt}_{1/8}(g)) is the {\em query partition bound} of the function gg. Partition bounds were introduced (in both communication complexity and query complexity models) to provide LP-based lower bounds for randomized communication complexity and randomized query complexity. Our results demonstrate that these lower bounds are polynomially tight for {\em product} distributions.

Cite

@article{arxiv.1512.01968,
  title  = {Partition bound is quadratically tight for product distributions},
  author = {Prahladh Harsha and Rahul Jain and Jaikumar Radhakrishnan},
  journal= {arXiv preprint arXiv:1512.01968},
  year   = {2020}
}

Comments

The previous version of the paper erroneously stated the main result in terms of relaxed partition number instead of partition number

R2 v1 2026-06-22T12:03:01.763Z