16.08.2018 Autoencoders and strong artificial intelligence (V. Morzhakov)
An article about training a set of autoencoders with a common latent space. An The context space implementation and its application for image analysis is shown.
05.07.2018 Implementation of reinforcement learning mechanisms in the context-semantic model of brain function (A. Redozubov)
Article about how the brain implements reinforcement learning. It shows how the idea of contexts can be used to create predictions. It is suggested that the brain uses V learning, that is, descriptions of the results of all possible actions are simulated in parallel. For each of the results, an estimate of the quality of the situation is calculated. Then the act which is promising the highest quality is being chosen. In this case, the role of predictive modules is performed by the minicolumn of the cortex, which implements the functions of context processors.
19.04.2018 Strong generalization. How the brain learns the essence of things (A. Redozubov)
Video lectures on the nature of strong generalization. The connection between concepts and contexts formed on their basis is shown.
16.12.2017 An Artificial Neural Network Architecture Based on Context Transformations in Cortical Minicolumns (V. Morzhakov, A. Redozubov)
An article on image recognition using the context space.
15.12.2017 Holographic Memory: A Novel Model of Information Processing by Neuronal Microcircuits (PDF) (A. Redozubov)
An article describing the memory mechanism and its relationship to the brain’s ability to work with contexts. It is shown that each mini-column of the cortex contains its own copy of the memory and is a “context processor”, that creates its own interpretation of the incoming information and determines how much this interpretation corresponds to previous experience. Each cortical zone is a space of appropriate contexts in which incoming information is treated simultaneously in all possible interpretations.
22.08.2016 Lecture series “Logic of Consciousness” (A. Redozubov)
Presentation of the current state of research. Description of the pattern-wave model of information transmission. Holographic distributed memory. Context-semantic model of brain cortex. The principles of spatial self-organization contexts. Visual system work. Training in combinatorial space. Examples of programs.
2011 Book “The Logic of Emotion” (A. Redozubov)
Pop-science book. The nature of emotions is described. It is shown that love, harmony, beauty and humor are different manifestations of a unified rational mechanism.