Researcher cracks new ‘kissing number’ bounds — besting AI in the process
How many coins can touch one coin, or how many basketballs can ‘kiss’ one basketball at the same time? This seemingly playful question lies at the heart of the famous kissing number problem, a mathematical riddle that becomes almost supernaturally difficult to work out in dimensions beyond 4D. Despite its whimsical name, similar problems have practical applications in areas such as mobile communications and satellite navigation.
Aalto University doctoral candidate Mikhail Ganzhinov established three new lower bounds for the kissing number: at least 510 in dimension 10, at least 592 in dimension 11, and at least 1,932 in dimension 14. There had been no movement on the riddle for dimensions below 16 for two decades until earlier this year, when AlphaEvolve, developed by Google's artificial intelligence laboratory DeepMind, made headlines in May. It was able to increase the lower bound of dimension 11 to a score of 593. So, only in the 11th dimension did Ganzhinov fall one step short of AlphaEvolve’s AI-powered result.
So how did the researcher beat the AI across the other two dimensions?
‘I reduced the problem size by looking only for arrangements with a high degree of symmetry,’ he explains. ‘In fact, the current lower bound for dimension 11 is still quite weak — I believe it can be pushed well beyond 600.’
Ganzhinov’s thesis advisor, Professor Patric Östergård, is impressed by the outcome — and quick to point out what it says about the limits of AI.
‘Artificial intelligence can do amazing things, but it’s far from omnipotent — and the game may still turn to Mikhail’s favour in dimension 11 too,’ Östergård remarks.
Recently awarded his PhD, Ganzhinov is modest about his achievements, noting that the field is evolving rapidly. Professor Henry Cohn from the MIT and researcher Anqi Li are set to publish new results that extend the kissing number bounds in dimensions 17 to 21 — the first progress in those dimensions in over 50 years. Ganzhinov says that his results are part of a broader wave of recent developments.
‘This riddle has challenged mathematicians since the famous conversation between Newton and Gregory,’ says Ganzhinov. ‘Yet solving them also has a practical purpose –– understanding connections to spherical codes has real life implications in the field of communications.’
Mikhael Ganzhinov's Doctoral thesis can be found here: Construction of few-angular spherical codes and line systems in Euclidean spaces.
This news item was originally published on the Aalto University website on 23.10.2025.
Read more news
Emotions help explain why people click phishing emails
People are currently taught to recognise phishing attempts in ways that may be counterproductive. Researchers call for information security training to take emotions into account more effectively.
New podcast explores Finnish contributions to artificial intelligence
In Aalto University's podcast “Ja ihminen loi älyn” (And humankind created intelligence), researchers Arno Solin ja Eric Malmi interview pioneers, entrepreneurs and experts about the AI field's history, disruptive consequences and the probability of humanity's destruction.
AI model captures how humans read, paving the way to personalised text and better augmented reality
Researchers now understand not just how our eyes move when we read, but also how we build meaning from text. The new model could hold the key to developing highly customisable applications that adapt to different types of reader.