In this video you have a small example how machine learning really looks.
NeuroBrain is a self-structuring and evolutionary Recursive Neural Network (SSERNN) developed by BolivarTech, it has the ability to define it internal topology by itself during the learning process, by aggregating new structures when it is needed in order to learn a knowledge or pattern, and removing the unused ones when are not needed; just like a biological brain.
Also it has the evolutionary support that allow the use of BolivarTech’s evolutionary algorithms implementation, where using the crossing between individuals and mutations through generations of NeuroBrains populations, allow search and find optimal and semi-optimal topologies that can learn in a more efficient way the knowledge; just like occurs in the natural environment as was described by Charles Darwin, where the more adapted to an environment pass it abilities to the next generations.
The use of the innovative Self-Structuring and the Evolutionary algorithms allow to NeuroBrains be a powerful deep machine learning solution.
In the example show in the video, is a population of NeuroBrains and over generations and generations perform TicTacToe match between them in order to learn and find strategies to win the game.
Several match are at the same time, everyone with their own competitors in order to use the CPU multi core support.
Julian Bolivar-Galeno is an Information and Communications Technologies (ICT) Architect whose expertise is in telecommunications, security and embedded systems. He works in BolivarTech focused on decision making, leadership, management and execution of projects oriented to develop strong security algorithms, Artificial Intelligence (AI) research and its applicability to smart solutions at mobile and embedded technologies, always producing resilient and innovative applications.
NeuroBrain, In a Small Introduction to Deep Machine Learning.
No comments:
Post a Comment