Two computer scientists at at the University of Liverpool think they have successfully cracked the Erdős discrepancy problem (for a particular discrepancy bound C=2), an 80 year old maths puzzle proposed by the Hungarian mathematician Paul Erdős, who offered $500 for its solution.
They just can't be sure, because it is too big for a human to replicate.The resulting proof generated is an enormous 13 gigabytes, 30 percent larger than downloading all of the content on Wikipedia.
Erdős was fascinated by the extent to which an infinite sequence of numbers containing nothing but +1s and -1s contains internal patterns. One way to measure that is to cut the infinite sequence off at a certain point, and then create finite sub-sequences within that part of the sequence, such as considering only every third number or every fourth.
Adding up the numbers in a sub-sequence gives a figure called the discrepancy, which acts as a measure of the structure of the sub-sequence and in turn the infinite sequence, as compared with a uniform ideal.
Dr. Boris Konev and Dr. Alexei Lisitsa took a sequence 1,161 numbers long and managed to demonstrate that an infinite sequence will always have a discrepancy larger than two. The resulting proof and its 13 gigabytes is probably the longest proof ever and dwarfs another famously huge proof, the Classification Theorem of Finite Groups, which totaled 15,000 pages.
But, although checking this is beyond the reach of humans, they suggest future developments could change that landscape.
Lisitsa, from the University’s School of Electrical Engineering, Electronics and Computer Science, said, “On the one hand, it is true that our computer-generated solution is beyond the reach of humans to fully understand. On the other, all we can say for now is that at the moment there is no known ‘better’ human-comprehensible solution – but it does not mean that such a solution could not (or will not) be found in the future.”
Spot a pattern
Konev added, “Erdős’ hypothesis was that a discrepancy of any value can always be found, a far cry from the discrepancies of 1 and 2 that have now been proven. Our software has been running for weeks in an attempt to find a result for discrepancy 3. But even if subsequent programs show that higher and higher discrepancies exist for any infinite sequence, a computer cannot check the infinity of all numbers.”
Instead, it is likely that computer-assisted proofs for specific discrepancies will eventually enable a human to spot a pattern and come up with a proof for all numbers, Lisitsa concluded.
Source: University of Liverpool.
- PHYSICAL SCIENCES
- EARTH SCIENCES
- LIFE SCIENCES
- SOCIAL SCIENCES
Subscribe to the newsletter
Stay in touch with the scientific world!
Know Science And Want To Write?
- Researchers Created A Laser Bullet To See What It Would Look Like - And Here It Is
- The Strange Organic Molecules In Titan's Atmosphere
- Will Holding Thermal Printer Paper Really Send Your BPA Levels Soaring?
- How Gut Bacteria Ensure A Healthy Brain – and Could Play A Role In Treating Depression
- The Quote Of The Week - Shocked And Disappointed
- As The Weather Changes, So Do Beliefs About Climate Change
- Moderate Pot Use By Adolescents Doesn't Hurt IQ
- "I see. One must have an expensive Mac to be able to use the software for free...."
- "Also, why do you assume a person seeking coverage needs to take off work? People in need of assistance..."
- "According to this article, the report was not about the accuracy of the website. It was a survey..."
- "Well how about you make some logically reasonable statements. My post was simply pointing out the..."
- "Okay, I understand your point now, but a human meeting another human is still a lot more costly..."
- US Ebola hysteria and money pit highlight lack of resources to confront diseases that kill far more people
- Addiction can be measured by epigenetics
- Coffee grounds turned biofuel can heat your home
- Bill and Melinda Gates on GMOs: ‘Poor farmers should not be denied choice of life-saving tools’
- Why do foodies love organics? Because they taste like McDonald’s!
- GMO milk? An enviros dream innovation that most enviros oppose
- Experimental breast cancer drug holds promise in combination therapy for Ewing sarcoma
- Desert streams: Deceptively simple
- UT Southwestern scientists discover new clues to how weight loss is regulated
- Gene that once aided survival in the Arctic found to have negative impact on health today
- Genomic data support early contact between Easter Island and Americas