Track 12

Brain Machine Interface

A brain-machine interface (BMI) is a device that converts neuronal data into commands accomplished of regulatory external software or hardware. The dream gadget that let humans read each other’s opinions and communicate with brain waves may be closer to reality. Brain-computer interfaces join knowledge and methods from neuroscience, signal processing, and machine learning. Functional Near Functional Infrared Spectroscopy (fNIRS) is used to record brain signals. The FNIRS is a non-invasive method to record brain indicators. Strengthening learning is an interactive learning method designed to permit systems to get a reward by learning to interact with the environment, and which has alteration built into the algorithm itself using a feedback signal.

Reinforcement learning (RL) algorithms can be separated into two categories model-free and model-based. Model-free studies a policy or value purpose. In model-free learning, the agent simply depends on some trial-and-error knowledge for action collection. In model-based learning, the agent deeds a previously learned lesson. However, these are based algorithms are dynamics model. Reinforcement learning is used for closed-loop brain-controlled interfaces.

Related Societies: Association for Computing Machinery (ACM), USA | British Automation and Robot Association (BARA), UK | Association Franchise pour Artificial Intelligence, France | Canadian Artificial Intelligence Canada | Japan Robot Association (JARA), Japan | International Federation of Robotics (IFR), Germany.