neuromorphic computing of the Human Brain Project

These chips can mimic important aspects of the biological brain through energy saving, flexibility, and learning ability. These chips are expected to have a major impact on the future of artificial intelligence.

Compared to the previous generation chip, the new BrainScale S chip has a high ability to reproduce the continuous learning process. Ideal for real-time simulation of a new generation of SpiNNaker chips with improved power management for highly efficient energy usage, multi-scale brain models

"We learned that learning is important to all applications of the neuromorphic system.The new architecture is optimized for rapid and effective learning.The idea of ​​common design process is today, It is the key to success and it is very proud of us all, "said Karlheinz Meier, who leads this field. Neuromorphic calculations perform aspects of biological neural networks as analog or digital copies on electronic circuits. The purpose of the neuromorphic calculation is to provide tools for neuroscience to understand the dynamic processes of brain learning and development and apply brain inspiration to general cognitive computing.

Two new chips were exhibited at the "Neero Inspirational Computing Components (NICE)" seminar held in the US at the end of February. Intel's neuromorphic chip was also exhibited.

Then there is the neuromorphic computing project of the European human brain project. It is working on building two large and unique neuromorphic machines and prototyping the next generation of neuromorphic chips. Recently, a paper demonstrating usefulness in computational neuroscience applications was released by completely simulating 80,000 neurons based on SpiNNaker hardware and 300 million synaptic cortical microcircuit models. Co-author of this paper, Professor Markus Diesmann, Director of Computational Systems Neurosciences at the Jülich Research Center in Germany, says: "There is a big gap between brain energy consumption and today's supercomputer.How much can you achieve energy efficiency of the brain using electronic equipment?"

A research project that includes neuromorphic engineering is the Human Brain Project. This is a ten-year collaboration that attempts to use biological data to simulate the entire human brain with a supercomputer. It consists of a team of researchers in the field of neuroscience, medicine and computation. Henry Markram, co-director of this project, said the project is building a foundation for investigating and understanding the brain and its diseases and suggesting building this new computing technology using this knowledge . The three main objectives of this project are to better understand how the various parts of the brain collaborate and cooperate, to understand ways to objectively diagnose and treat brain diseases, It is to utilize human brain's understanding to develop Morphic computer. To simulate a perfect human brain, you need a super computer that is 1000 times more powerful than today. This attracts the current attention to neuromorphic computers.

Let's clarify that the human brain project is huge. There are a lot of excellent competent scientists and engineers working on high quality. There is a complete project dedicated to the "neuromorphic" chip which uses model neurons instead of binary transistors. To test the limits of our understanding there is a commitment to materializing robotics where brain models are tested in a real environment. People's ability to artificial intelligence is now increasingly hysterical, so this is a large part of ethics and is very popular. The core model itself is a technical miracle. To build it we need to gather a large amount of data on neuron's detailed shape, size, and routing, as well as all electrical activity measurements. To synthesize all the data into the model, we need to develop many new methods to estimate using experimental data.