qiskit-camp-asia-19 icon indicating copy to clipboard operation
qiskit-camp-asia-19 copied to clipboard

Implementing quantum multi-collision algorithm on Qiskit

Open ace314 opened this issue 5 years ago • 13 comments

Abstract

Finding collisions for a many to one mapping is a fundamental problem in cryptography. In this project, we would implement quantum multi-collision algorithm on Qiskit. Rather then simply applying Grover search, which gives a quadratically speed up, we would like to realize more efficient algorithms to attack the multi-collision problem.

Description

The multi-collision problem states as follows: For given finite sets and with   , and a function , the s-collision finding problem is to find a set of distinct inputs such that

.

For s=2 cases, the problem corresponds to the well-known "birthday paradox", which shows that the probability to find two people sharing the same birthday in a banquet is much more larger than one may intuitively think. In fact, the probability reaches 50% with only 23 people in a room. One can use Grover search to solve the multi-collision problem for     expected evaluations. Moreover, the optimized algorithm so called "BHT algorithm" gives     for s=2 and any arbitrary function (arXiv:quant-ph/9705002). Recently, some algorithms have been proposed with higher efficiency or for more complicated conditions such as s>2 and different features of (arXiv:1911.02822). In this project, we would look into the algorithms already proposed and try to do the implementation on Qiskit. Then based on that, we optimize the algorithm as far as possible during the hackathon.

Members

  • @abtmy - Slack: @Tomoya Abe
  • @r1t0o0 - Slack: @Rito
  • @ace314 - Slack: @Che Chiang
  • @BOBO1997
  • @starktech23
  • @githubhandle - Slack: @slackhandle email: [email protected]
  • Qiskit Coach: @atilag

Deliverable

GitHub repo

ace314 avatar Nov 15 '19 05:11 ace314

I'm very interesting in this topic. I've finish reading the papers and had some thoughts. Maybe we can discuss some ideas through email?

R07222062 avatar Nov 17 '19 11:11 R07222062

I'm very interesting in this topic. I've finish reading the papers and had some thoughts. Maybe we can discuss some ideas through email?

Tnx for your noticing! Feel free to email me at any time. It would be nice if you could post your thoughts here :)

ace314 avatar Nov 17 '19 13:11 ace314

I'm interested in this project. This algorithm is used when we consider post-quantum security of hash functions.

abtmy avatar Nov 17 '19 16:11 abtmy

I'm interested in this project. This algorithm is used when we consider post-quantum security of hash functions.

I am glad to here that. Maybe we can discuss the details tomorrow. See you then :)

ace314 avatar Nov 17 '19 17:11 ace314

It sounds interesting to me, count me in

r1t0o0 avatar Nov 19 '19 01:11 r1t0o0

@1ucian0 please assign me to this group

r1t0o0 avatar Nov 19 '19 02:11 r1t0o0

@1ucian0 please add me to this group, thank you!

BOBO1997 avatar Nov 19 '19 02:11 BOBO1997

interested. Samanvay Sharma, Computer Science.

starktech23 avatar Nov 19 '19 02:11 starktech23

Masahiro Fujii, Computer science

mf-22 avatar Nov 19 '19 02:11 mf-22

BOBO1997, Yang Bo, Computer Science

BOBO1997 avatar Nov 19 '19 02:11 BOBO1997

Yu GUO, Computer Science, familiar with classical algorithm

r1t0o0 avatar Nov 19 '19 02:11 r1t0o0

abtmy, Tomoya Abe, computer science

abtmy avatar Nov 19 '19 02:11 abtmy

https://github.com/ace314/Quantum-Multi-collision-Problem

ace314 avatar Nov 20 '19 05:11 ace314