Dublin Institute for Advanced Studies contact@dias.ie 00353 (0) 16140100

DIAS Team Takes First Place in IBM’s Quantum Computer Hackathon

Starting on 18th August, teams from across Ireland went head-to-head in the IBM Qiskit Quantum Summer Jam Hackathon. The competition was open to the public, so naturally the 13 teams covered a diverse array of backgrounds, from universities and research institutes, to companies like Mastercard. Teams had only a single week to put together a programming project of their choosing using IBM’s cutting edge quantum computer. At the end of the week the teams submitted their open source code (available on GitHub) and a presentation describing their project. It was then down to the judges to decide which project best showcased IBM’s quantum computer. But what exactly is a quantum computer, and why are researchers from across Ireland (and the world in other national IBM competitions) excited to use IBM’s machines?

Quantum computers, unlike everyday computers, take advantage of quantum phenomena such as superposition and entanglement. While these are complicated concepts, the important point is that quantum computers have the potential to far exceed the storage and computation capacities of everyday computers, like the one you may be using now. This comes at a price of course. Quantum phenomena, and hence quantum computers, are inherently random! Imagine a computer that sometimes alters your saved files; this is obviously not a desirable feature. That being said, this isn’t always an issue with quantum computers if we are able to control the randomness to some desired accuracy.

Still looming, however, is the practical issue that quantum computers are just plain difficult to make. IBM and Google’s quantum computers are highly tuned delicate machines taking up entire rooms! But still they only have a handful of qubits (the quantum analogue of computer bits). For an analogy, think of the birth of the modern computer; Alan Turing building a room-sized computer to crack the WWII Enigma machine in the 1940s.

Seen in this way, it is clear why computer scientists and theoretical physics are jumping at the opportunity to use IBM’s open access quantum computer. It’s a chance to be a small part of history!

As well as the hardware questions raised above, there are also questions of how best to use quantum computers. In programming terms, what algorithms work best on quantum computers? In IBM’s hackathon, competitors had to implement and fine tune existing algorithms, or even invent entirely new ones! The runner’s up, for example, implemented an existing algorithm for solving a 2×2 Rubik’s cube. The winning team from DIAS — led by Luuk Coopmans (Ph.D. student with DIAS and Trinity), with team members Ian Jubb (Postdoctoral Fellow at DIAS), DIAS summer students Cillian Doherty (Trinity) and Maria Graham (UCD), and Rajarshi Tiwari (Postdoc at Trinity) — implemented and improved upon novel image processing and watermarking algorithms. In simple terms, they were able to save and retrieve images to and from IBM’s quantum computer, and add secret watermark images only decode-able by the desired recipient.

As an example, here are a couple images saved to and retrieved from the quantum computer:

The DIAS logo.
A cartoon of Schrodinger’s cat (rest assured, in the
age of quantum computers we’ll still be able to view cat pictures on the
internet).

The recovered images are a little fuzzy, but saving and retrieving images to and from the quantum computer is not the selling point of what the DIAS team achieved. The real selling point is the speed at which the algorithm encrypts and decrypts the secret watermark. Essentially, the algorithm allows for multiple pixels of an image to be altered at the same time, whereas an everyday computer can only alter one at a time. The image encryption used in this type of watermarking is built upon pixel alteration, and hence the algorithm can encrypt/decrypt images much faster than an everyday computer. This is just one particular application of a more general property of quantum algorithms — quantum parallelism. One of the more famous applications of this property is Shor’s algorithm for breaking RSA encryption, which is one of the most commonly used methods to send data securely. The somewhat terrifying prospect that quantum computers can break RSA encryption much faster than everyday computers has certainly piqued the interests, and concerns, of researchers from around the world. 

IBM’s competition and global security concerns aside, this is still only the dawn of the age of quantum computers. Nevertheless, the DIAS team were happy to get their spot for the sunrise.