Computing: Towards the unconventional convention

New Zealand is to play host to the first-ever international conference on unconventional models of computation in January next year. It is the first time there has been a conference combining quantum, molecular and other computing models

New Zealand is to play host to the first-ever international conference on unconventional models of computation in January next year.

It is the first time there has been a conference combining quantum, molecular and other computing models.

The conference has been organised by the Santa Fe Institute and the joint Auckland-Waikato University Centre for Discrete Mathematics and Theoretical Computer Science (CDMTCS).

The organisers have invited some high- profile experts in theoretical computing to speak at the conference, including Nobel Prize winner Ilya Prigogine of the Solvay Institute and scientists from Oxford, MIT and Caltech.

“It has been hard getting everybody in one place at one time, so we started organising this two years ago,” says conference committee chair Cristian Calude.

Calude says there are basically two sides to unconventional computing. The first is that there may be a theoretical limit to the things it is possible to compute – just as you can only fold a piece of paper in half, then in half again six times.

“This is not a maths theorum but a hypothesis which has been checked and verified many times over the years. Any non-conventional theory is one which suggests a way to go beyond this limit.”

The second side is that if we are indeed confined to what is known as the “natural range” of computations then it should be possible to speed up computations to get more out of computers.

“There is a theory we are not using conventional computers to their capabilities because we don’t have the appropriate notation for very large numbers,” says Calude.

He likens the concept to performing arithmetic with the Roman numeral system versus the Arabic number system we use today.

“They are both valid number systems but the Roman numeral system is very hard to use for things like multiplication. It is like we are using the equivalent of Roman numerals in computing and this inhibits speed.”

Calude says in the field of unconventional computing the idea is to make use of different opportunities offered by combining computing with other sciences such as quantum and biological studies.

Conference secretary Michael Dinneen says an unconventional model of computing tends to be on a smaller scale.

“For example, electrical circuits are used in the conventional model but unconventional computing would use molecule computations, and quantum would be even more miniature.”

Calude says computers as we know them obey the laws of classical mechanics – a computer in an office in New Zealand cannot be in New York at the same time.

“But in quantum physics, one part of an atom can be in many places. If a computer was made up of a quantum molecule then a transaction could run in parallel in many places. The idea is to speed up computation by doing things in parallel. The same with molecular computing – you are looking globally at many molecules at the same time, you’re looking at the forest not the trees.”

At the moment there are few practical examples of these theories being carried out in the laboratories around the world, but some work is being done.

Some laboratories have done molecular computing by combining tubes of DNA material. When this is sorted in the gene mapping process a string of information is produced. This can then be matched with other DNA strings to see if they are similar – thus a lot of information can be processed in one step rather than working through a large number of transactions, the way a computer would.

“The biological model has already been found to be more efficient by researchers in the US who were working on the travelling salesman problem – where a salesman has a string of cities to visit and must determine which route, taking in all cities, is the quickest. The biological computer found the solution in one step by combining tubes together by hand. In future this could be done by robots,” says Dinneen.

While the conference will have a couple of people who are doing practical experiments in the unconventional computing area, it will be mostly theory.

“Computers started making their name in the 1930s,”says Calude, “Non-conventional computing is today what computers were back in the 1930s. We need dreamers – this kind of thing is on the border between science and science fiction. There are people in the laboratory actually trying to implement the theories but we need one team ahead and the other team one or two steps behind coming along and telling those in front what you can and can’t practically do.”

Cristian says the conference will help make New Zealand a little more visible internationally in an area not covered much by other universities.

“There is a slim possibility something may come out of it at a commercial level, but it is very important for our students to be exposed to a number of eminent people from around the world.”

Dinneen agrees: “That is one of the most important aims – most computer students don’t know what research is – they just do programming, so every day there will be times for students to mingle with the speakers and participate in discussions.”

For more information see the conference home page at http://www.cs.auckland.ac.nz/CDMTCS/docs/news.html.

Join the newsletter!

Error: Please check your email address.
Show Comments
[]