Good afternoon. I am developing my own original approach to designing Strong Artificial Intelligence.
The core idea of my approach is that human intelligence itself has its own multi-level structure (not to be confused with the structure of the brain). Each level of perception is responsible for processing external signals of a specific type, from the simplest signals such as sound, smell, touch, retinal images, etc., to complex signals such as speech, scientific theories, works of art, etc. I assume that a neural network analogue of human intelligence will be able to approach the functionality of human intelligence if it replicates the structure of human intelligence. Only then a neural network analogue of human intelligence will be able to perform intellectual work in the same way as human intelligence. To do this, we need to formulate the structure of human intelligence (not the brain, but the cognitive apparatus itself). I propose and describe this structure in my article [https://mtnip.elpub.ru/jour/article/view/35] In this article, I also propose an algorithm for training a neural network analogue of human intelligence at each individual level and the entire predictor using tools consisting of generative networks and genetic algorithms. I would like to understand whether my proposed approach might be of interest to You, DeepSeek and if so, how I could participate in Your development using the proposed approach.
I am ready to present a detailed program for the development and implementation of a neural network predictor based on the proposed approach, should You be interested. I would be very grateful for any response to my email phaust@mail.ru.
Good afternoon. I am developing my own original approach to designing Strong Artificial Intelligence.
The core idea of my approach is that human intelligence itself has its own multi-level structure (not to be confused with the structure of the brain). Each level of perception is responsible for processing external signals of a specific type, from the simplest signals such as sound, smell, touch, retinal images, etc., to complex signals such as speech, scientific theories, works of art, etc. I assume that a neural network analogue of human intelligence will be able to approach the functionality of human intelligence if it replicates the structure of human intelligence. Only then a neural network analogue of human intelligence will be able to perform intellectual work in the same way as human intelligence. To do this, we need to formulate the structure of human intelligence (not the brain, but the cognitive apparatus itself). I propose and describe this structure in my article [https://mtnip.elpub.ru/jour/article/view/35] In this article, I also propose an algorithm for training a neural network analogue of human intelligence at each individual level and the entire predictor using tools consisting of generative networks and genetic algorithms. I would like to understand whether my proposed approach might be of interest to You, DeepSeek and if so, how I could participate in Your development using the proposed approach.
I am ready to present a detailed program for the development and implementation of a neural network predictor based on the proposed approach, should You be interested. I would be very grateful for any response to my email phaust@mail.ru.