The problem of aging as a problem of the appearance of unstable components when detailing the system

Abstract

Recently, interest in Physical Informed Neural Network (PINN) has grown. But the process of implementing these models in the field of mathematical biology is still in its infancy. For this area, the question of finding a region of stability is both very difficult and important. There are many models for disease, aging, and carcinogenesis, but they all rely on the concept of an attractor of human health and resilience. But if you teach Artificial Intelligence to maintain the region of stability of a healthy state, it will inevitably face entering the region of instability, in which there is great sensitivity to changes in parameters, which will give exponentially large gradients. But when it comes to critical processes, organisms have a fairly good set of ways to maintain short-term stability that they have acquired through the process of evolutionary selection. Thus, all factors that affect aging are dynamics with slow exponential growth throughout life, similar to DNA mutations. Therefore, a hypothetical rough model that could directly suppress all such deep problems in their infancy may be an interesting first approximation, although most likely, due to excessive suppression, it will simplify the structure of the organism itself and possibly make its tissues and coordination of cellular reprogramming more similar to the tissues of the immortal jellyfish Turritopsis dohrnii. This abstract approximation model has long-term stability, and from there we can view aging as a problem of unstable components emerging in the detailing of the organism that we must solve.