Nano artificial neural network is expected to create "artificial brain"
Release date: 2015-05-15 With the development of artificial circuits that simulate the human brain, super-intelligent robots may soon become a reality. For the first time, scientists have created 100 human synapses capable of performing human image recognition tasks. Scientists say their nano-memory cells are 10,000 times thinner than human hair and can store and process chain information at the same time. Nano-memory cells mimic the natural state of the human brain by preserving different states of the data. Dmitri Strukov, professor of electrical and computer engineering at Santa Barbara, said that although this is a small step, it is an important step. The circuit may eventually expand and reach the scale of the human brain, with about one trillion synaptic links in the human brain. Simulating the electrical properties of the human brain may contribute to the study of Alzheimer's disease and Parkinson's disease. Scientists believe that if we can replicate the human brain in vitro, we can avoid the ethical problems caused by human brain experiments. The latest results show that artificial neural network circuits can successfully classify "z", "v", and "n". It can analyze the image of letters, each letter has a different signal noise. It is similar to human beings who distinguish their friends from the crowd. The key to solving this problem is to remember the impedance, which is a combination of memory and resistance. It is an electrical component that relies on the direction of the current to change the resistance. Resident impedance is not like traditional transistors, which rely on ion motion to work, similar to the principle that human brain cells produce electrical signals. Potential applications for this technology already exist, such as medical images, navigation system improvements or image-based searches. At the same time, researchers are also working to improve the performance of the memory and expand the complexity of the circuit. The next step is to combine the impedance neural network with traditional semiconductor technology. This will enable the creation of more complex versions that will allow early artificial brains to do more complex and detailed work. Ideally, this "brain" trillion synapse can vertically integrate information between each other. Source: Kexun Medical Network
Offline One Time Code (OTC) Lock to secure the assets in Safe or Vault or ATM such financial facilities.
Using bank grade encryption technology, easy to use APP, can supports two codes to combination in the Lock to enhance the security of the password.
One time code make sure that the pasword will get expired one used.
To assign the task with management system can make sure that each unlock operation is tracable.
Smart lock, OTC lock Shenzhen Sunson Tech Co., Ltd , https://www.sunsonkiosk.com