Near-optimal lower bounds on the multi-party communication complexity of set disjointness

Amit Chakrabarti, Subhash Khot, Xiaodong Sun

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We study the communication complexity of the set disjointness problem in the general multi-party model. For t players, each holding a subset of a universe of size n, we establish a near-optimal lower bound of Ω(n/(t log t)) on the communication complexity of the problem of determining whether their sets are disjoint. In the more restrictive one-way communication model, in which the players are required to speak in a predetermined order, we improve our bound to an optimal Ω(n/t). These results improve upon the earlier bounds of Ω(n/t2) in the general model, and Ω(ε 2n/t1+ε) in the one-way model, due to Bar-Yossef, Jayram, Kumar, and Sivakumar [5]. As in the case of earlier results, our bounds apply to the unique intersection promise problem. This communication problem is known to have connections with the space complexity of approximating frequency moments in the data stream model. Our results lead to an improved space complexity lower bound of Ω(n1-2/k/log n) for approximating the kth frequency moment with a constant number of passes over the input, and a technical improvement to Ω(n1-2/k) if only one pass over the input is permitted. Our proofs rely on the information theoretic direct sum decomposition paradigm of Bar-Yossef et al [5]. Our improvements stem from novel analytical techniques, as opposed to earlier techniques based on Hellinger and related distances, for estimating the information cost of protocols for one-bit functions.

Original languageEnglish (US)
Title of host publicationProceedings of the Annual IEEE Conference on Computational Complexity
Pages107-117
Number of pages11
StatePublished - 2003
Event18th Annual IEEE Conference on Computational Complexity - Aarhus, Denmark
Duration: Jul 7 2003Jul 10 2003

Other

Other18th Annual IEEE Conference on Computational Complexity
CountryDenmark
CityAarhus
Period7/7/037/10/03

Fingerprint

Communication Complexity
Lower bound
Communication
Space Complexity
Moment
Model
Data Streams
Direct Sum
Disjoint
Intersection
Paradigm
Decomposition
Decompose
Subset
Costs

ASJC Scopus subject areas

  • Computational Mathematics

Cite this

Chakrabarti, A., Khot, S., & Sun, X. (2003). Near-optimal lower bounds on the multi-party communication complexity of set disjointness. In Proceedings of the Annual IEEE Conference on Computational Complexity (pp. 107-117)

Near-optimal lower bounds on the multi-party communication complexity of set disjointness. / Chakrabarti, Amit; Khot, Subhash; Sun, Xiaodong.

Proceedings of the Annual IEEE Conference on Computational Complexity. 2003. p. 107-117.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chakrabarti, A, Khot, S & Sun, X 2003, Near-optimal lower bounds on the multi-party communication complexity of set disjointness. in Proceedings of the Annual IEEE Conference on Computational Complexity. pp. 107-117, 18th Annual IEEE Conference on Computational Complexity, Aarhus, Denmark, 7/7/03.
Chakrabarti A, Khot S, Sun X. Near-optimal lower bounds on the multi-party communication complexity of set disjointness. In Proceedings of the Annual IEEE Conference on Computational Complexity. 2003. p. 107-117
Chakrabarti, Amit ; Khot, Subhash ; Sun, Xiaodong. / Near-optimal lower bounds on the multi-party communication complexity of set disjointness. Proceedings of the Annual IEEE Conference on Computational Complexity. 2003. pp. 107-117
@inproceedings{3aeccbdbc4af4f15af1a31516c46f783,
title = "Near-optimal lower bounds on the multi-party communication complexity of set disjointness",
abstract = "We study the communication complexity of the set disjointness problem in the general multi-party model. For t players, each holding a subset of a universe of size n, we establish a near-optimal lower bound of Ω(n/(t log t)) on the communication complexity of the problem of determining whether their sets are disjoint. In the more restrictive one-way communication model, in which the players are required to speak in a predetermined order, we improve our bound to an optimal Ω(n/t). These results improve upon the earlier bounds of Ω(n/t2) in the general model, and Ω(ε 2n/t1+ε) in the one-way model, due to Bar-Yossef, Jayram, Kumar, and Sivakumar [5]. As in the case of earlier results, our bounds apply to the unique intersection promise problem. This communication problem is known to have connections with the space complexity of approximating frequency moments in the data stream model. Our results lead to an improved space complexity lower bound of Ω(n1-2/k/log n) for approximating the kth frequency moment with a constant number of passes over the input, and a technical improvement to Ω(n1-2/k) if only one pass over the input is permitted. Our proofs rely on the information theoretic direct sum decomposition paradigm of Bar-Yossef et al [5]. Our improvements stem from novel analytical techniques, as opposed to earlier techniques based on Hellinger and related distances, for estimating the information cost of protocols for one-bit functions.",
author = "Amit Chakrabarti and Subhash Khot and Xiaodong Sun",
year = "2003",
language = "English (US)",
pages = "107--117",
booktitle = "Proceedings of the Annual IEEE Conference on Computational Complexity",

}

TY - GEN

T1 - Near-optimal lower bounds on the multi-party communication complexity of set disjointness

AU - Chakrabarti, Amit

AU - Khot, Subhash

AU - Sun, Xiaodong

PY - 2003

Y1 - 2003

N2 - We study the communication complexity of the set disjointness problem in the general multi-party model. For t players, each holding a subset of a universe of size n, we establish a near-optimal lower bound of Ω(n/(t log t)) on the communication complexity of the problem of determining whether their sets are disjoint. In the more restrictive one-way communication model, in which the players are required to speak in a predetermined order, we improve our bound to an optimal Ω(n/t). These results improve upon the earlier bounds of Ω(n/t2) in the general model, and Ω(ε 2n/t1+ε) in the one-way model, due to Bar-Yossef, Jayram, Kumar, and Sivakumar [5]. As in the case of earlier results, our bounds apply to the unique intersection promise problem. This communication problem is known to have connections with the space complexity of approximating frequency moments in the data stream model. Our results lead to an improved space complexity lower bound of Ω(n1-2/k/log n) for approximating the kth frequency moment with a constant number of passes over the input, and a technical improvement to Ω(n1-2/k) if only one pass over the input is permitted. Our proofs rely on the information theoretic direct sum decomposition paradigm of Bar-Yossef et al [5]. Our improvements stem from novel analytical techniques, as opposed to earlier techniques based on Hellinger and related distances, for estimating the information cost of protocols for one-bit functions.

AB - We study the communication complexity of the set disjointness problem in the general multi-party model. For t players, each holding a subset of a universe of size n, we establish a near-optimal lower bound of Ω(n/(t log t)) on the communication complexity of the problem of determining whether their sets are disjoint. In the more restrictive one-way communication model, in which the players are required to speak in a predetermined order, we improve our bound to an optimal Ω(n/t). These results improve upon the earlier bounds of Ω(n/t2) in the general model, and Ω(ε 2n/t1+ε) in the one-way model, due to Bar-Yossef, Jayram, Kumar, and Sivakumar [5]. As in the case of earlier results, our bounds apply to the unique intersection promise problem. This communication problem is known to have connections with the space complexity of approximating frequency moments in the data stream model. Our results lead to an improved space complexity lower bound of Ω(n1-2/k/log n) for approximating the kth frequency moment with a constant number of passes over the input, and a technical improvement to Ω(n1-2/k) if only one pass over the input is permitted. Our proofs rely on the information theoretic direct sum decomposition paradigm of Bar-Yossef et al [5]. Our improvements stem from novel analytical techniques, as opposed to earlier techniques based on Hellinger and related distances, for estimating the information cost of protocols for one-bit functions.

UR - http://www.scopus.com/inward/record.url?scp=0041513338&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0041513338&partnerID=8YFLogxK

M3 - Conference contribution

SP - 107

EP - 117

BT - Proceedings of the Annual IEEE Conference on Computational Complexity

ER -