Three math students of the University of Salento played the numbers of the lot suggested by an artificial intelligence system. They created it based on the numbers released in the last draws. The idea of being able to predict the draws is fascinating but unfortunately it cannot be applied for us.
The story of the boys who have used artificial intelligence for the lot is interesting. In short. Three students from the University of Salento have tried to use artificial intelligence models to predict the lotto numbers. They played those numbers and won. The three students are students of the Faculty of Mathematics And they haven’t simply asked Chatgpt how to do but they created one of their model.
The winnings were two: of 4,500 and 43,000 euros. Both were recorded a Monteroni di Lecce. The news has no more details but obviously has raised a lot of interest: the three boys found the magic formula To go beyond the layout of the lot numbers?
The system that used the three boys
Curious how all this story has gods not very clear contours. There is no name of the boys who played and there is no evidence that we used a system generated by artificial intelligence. The only testimony is that of the tobacconist who recorded the games, Diego Manca. From what you understand from his statements, the boys have decided to Analyze the most recurring numbers released in the last period in the draws of the lot on the national wheel. Studying the frequency of numbers they have chosen those on which aim for winning games.
Our phones become virtual assistants: they are useful but there is a hidden price that we are paying
Let’s do a practical example with chatgpt. According to the chatbot in the last three months, the number 38 has been extracted 22 times in the national lotto wheel. The number 10 came out 21 times. The number 1 has been extracted 20 times, and so on. From this base we hypothesize that the three students will have decided how to combine the numbers of their games.
What is needed to overcome the draws of the lot
Was it enough so little? Obviously not. The system has no memorydoes not follow patterns, there are no models that return. Of course, patterns can be identified in draws, looking for an order in chaos but it is a totally arbitrary exercise. At each game the numbers are the same probability of exiting the previous game.
He explains it well on the pages of Corriere della Sera Fabrizio duringprofessor of probability and mathematical statistics right at the University of Salento attended by students: “The game of lot is purely random, therefore it is not possible to predict the outcome of an extraction based on the previous ones. Also strategies based on numbers already extracted They don’t give the certainty of winning. In the long run, the gain is null or a loss. “
And again: “I don’t know in detail the method they used, but we know that in the lot there is no winning strategy in the long run. They can still win, of course, but they can also lose for the next two thousand games“.
The paradox is precisely this. As we had explained in the interview with Nicola Parolini, professor of the Department of Mathematics of the Polytechnic of Milan, from a statistical point of view The thing that is better to do with the lot is not to play. In the end the only factor that matters is luck.
