Mathematicians created. A new way to predict the future.

“ An international team of mathematicians has developed a new predictive method called the Maximum Agreement Linear Predictor (MALP). Unlike traditional techniques that focus only on minimizing average errors, MALP optimizes the Concordance Correlation Coefficient (CCC) to maximize agreement between predictions and actual outcomes. Credit: Shutterstock” (ScitechDaily, Mathematicians Unveil a Smarter Way to Predict the Future)

“In statistics, the concordance correlation coefficient (CCC) measures the agreement between two variables, e.g., to evaluate reproducibility or for inter-rater reliability.” (Wikipedia, Concordance correlation coefficient)

“The team evaluated MALP using both computer simulations and real-world data, including eye scans and body fat measurements. To demonstrate its effectiveness, the researchers applied MALP to data from an ophthalmology study comparing two optical coherence tomography (OCT) devices: the older Stratus OCT and the newer Cirrus OCT. Because clinics are shifting to the Cirrus system, physicians need a reliable way to convert measurements to ensure consistency over time and across devices.” (ScitechDaily, Mathematicians Unveil a Smarter Way to Predict the Future)

“Optical coherence tomography (OCT) is a high-resolution imaging technique with most of its applications in medicine and biology. OCT uses coherent near-infrared light to obtain micrometer-level depth resolved images of biological tissue or other scattering media. It uses interferometry techniques to detect the amplitude and time-of-flight of reflected light.”(Wikipedia, Optical coherence tomography)

There is a possibility of using the methods. That meant for OCT data processing to process data from the other sources. In those cases. Regular cameras and other sensors transmit data to the data-processing unit. That was originally created for the OCT. The sensor that transmits data is not important, and the software can make multiple analyses. From astronomy to sociology. In those other cases, the data portals that the system handles must only be renamed. The system must consider that the data it gets is not original medical data, and that’s why those systems must be calibrated for their other missions. 

The thing is that every single program. That can handle transformation in systems. It can predict the future. Like how the system grows. It can be used to predict. How all kinds of systems behave. If we want to transform systems that predict how the cancer grows. To predict how corruption behaves in socio-economic situations. Researchers must rename things. In those systems, the body can be the society. Immune cells can be law enforcement. etc. 

That can cause cancer. To actors who boost corruption. Things like carcinogens can be money that feeds corruption. Cancer cells can turn into corrupted officials. And free radicals. Are things. Like drug money. That thing can be used to predict how corruption spreads in the system. Of course, things like psychological effects. It can have a vital role in those things. The body that is polluted with carcinogens can be society. That accepts bibes. 

The digital twin of an average human can be used to create models. About how humans behave in certain situations. Those things are important when we try to predict the future. Like wars and peace. There are many things. That can cause war. And one of them is. The collapse in irrigation or food production. The other reason can be that. The ruler wants an outside enemy. There, the dictator can turn people's focus. When they start to call for new elections. 

The mathematical model to predict the future is quite simple. The system must just compile things that happened in the past. Then the system must collect all data about the natural, sociological, and political environments. And then the system predicts those values in a modern environment. This idea is taken from psychological models. That all humans behave in similar ways. In similar situations. There are always exceptions to the common models. Those exceptions are people who behave differently from the so-called average or standard person. Those people are people who behave differently, like schizophrenia patients. But for creating the mathematical version of average humans. 

The system must know how the majority of people behave. And the system uses that data. To create an average person’s digital twin. And then the system follows how the average person behaves using that model. There are also other types of exceptions. And one is people who have a very large media audience. Those people can escalate their opinions into a very large area. Those large-scale exceptions are the reasons why making predictions of the future is so difficult. 

Prediction of the future means. The system connects databases together, and then. It starts to find similarities in the modern world. “This new approach aims to generate predictions that align more closely with actual outcomes. The researchers call it the Maximum Agreement Linear Predictor, or MALP. The method achieves higher consistency by optimizing the Concordance Correlation Coefficient (CCC), a metric that evaluates how well pairs of data points align along the 45-degree line of a scatter plot.” (ScitechDaily, Mathematicians Unveil a Smarter Way to Predict the Future)

https://scitechdaily.com/mathematicians-unveil-a-smarter-way-to-predict-the-future/

https://en.wikipedia.org/wiki/Concordance_correlation_coefficient

https://en.wikipedia.org/wiki/Optical_coherence_tomography

What if we put computers to think mathematically?

When we think about programming and the computer's memories every single memory unit in the computer hardware has a certain address. The artificial intelligence connects and disconnects those memory points into the new orders. And that can make a computing process that mimics thinking. Every single memory address is like a piece of the puzzle. If every single memory unit has a certain number that makes it possible to point certain points from the computer memory. 

When we think about things like thinking we could easily connect those memory units with orders that the large language model, LLM gets by using the numeric values of the memory units and then calculate them with the ASCII marks. In the ASCII system, every single mark on the keyboard has a numeric value. 

For example, the letter A has a numeric value 61 in decimal and 41 in hex. A little a (a) has values 97 in decimal and 61 in hex. That's why it's not the same as the letter big or small in passwords. The numeric system is also important. The hexadecimal ("Base-16" system where the 10 comes after 16) and regular decimals are different. 

In that system 10 is marked in the numeric line like this. 0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F,10. In binary system 10 comes after 9 like this. 0,1,2,3,4,5,6,7,8,9,10.

Same way every single color has a numeric form in the computer memory. The system is known as RGB.  The system can use CCD cameras to make observations. 

Another thing is to use the values that fit to computer or programmer better. The color red can have a numeric value "200" and then the depth of that color can have 99 states. The system can turn every color into its own numeric value. 

The deepest red can be the 299. The data that CCD camera pixels give can be numeric. The system can see what numeric value every pixel gives and then it can make the model about things that it sees. So all data that travels into the system can turn into numeric. 

Token ring. If we think of this model as the computing cycle of the AI.  The system connects data into that data cycle. Every point in the cycle. There is the computer's image.

This can be the new way to handle large language models, LLMs are not to turn their mathematical models for words. The system can translate data that users input there into the mathematical model. Then the LLM starts to operate and process data in the mathematical form. That kind of thing can be lighter for computers than the words that we use. Mathematics is easier for computers, and when we think about the ability to turn words into mathematical form, we must remember that ASCII codes are basically numbers. Those numbers can sum, division, and multiplicate easier than words. 

That means the LLM can turn every single word that it has into numbers. Then that system can make calculations using the numbers. The ability to handle data in numeric form makes those systems more effective. The system can use the "token ring" type data handling, or computing model. The token ring model is known from data networks. However, the same model can introduce how the system surrounds data in it. Every time, when the system makes the data cycle it connects information into that data cycle. 

The system makes a certain number of calculations in every round. In those calculations, the system connects data from the sensors and memories in the data flow. The system doesn't need to show that information to the users before it drives it through the cycle as many times as ordered.

https://www.geeksforgeeks.org/ascii-table/

https://www.quantamagazine.org/to-make-language-models-work-better-researchers-sidestep-language-20250414/

https://www.rapidtables.com/web/color/RGB_Color.html

https://en.wikipedia.org/wiki/Hexadecimal

https://en.wikipedia.org/wiki/RGB_color_model