Thursday, February 22, 2018

Analysis of neural computation might have application for deep learning




The cortical brain is an evolutionary marvel which interacts with the outside world via self-regulation, based on its resting or ground state. The resting state, which is disturbed during the stimulus and sensory processing, is recovered by automatic operations. A new research articleTHE THERMODYNAMIC ANALYSIS OF NEURAL COMPUTATION, examines the brain's sensory processing as an energy-information cycle. The work also has relevance for artificial intelligence, such as deep learning. 

The brain’s energy need multiplies in warm-blooded animals. Although the human brain represents only 2% of the body weight, it receives 15% of the cardiac output, 20% of total body oxygen consumption, and 25% of overall body glucose utilization. The baseline energy consumption might be entirely dedicated to neuronal signaling. This constant and massive energy use maintains the brain’s alertness, even during sleep, by a dynamic balance between inhibitory and excitatory neurons. 

The slightest variation in excitation determines whether a spike is generated. Such a delicate balance of excitatory and inhibitory neurons turns the resting state into a highly energy-requiring state. Even in the absence of stimulus, the delicate balancing of excitatory and inhibitory neurons produces seemingly arbitrary activations. Like the tennis player's anticipation of the opponent's serve, the brain’s dynamic balance ensures its ability for a rapid, targeted response. For example, changes in inhibitory neurons increase frequencies and their energetic needs. The brain partakes in the energy information exchange with the environment via the sensory system.



Sign up to receive notification about the publication of my Upcoming book on emotion regulation





The Science of Consciousness Post, your news about the mind
The Science of Consciousness, please join the discussion
Website: evadeli
Copyright © 2017 by Eva Deli

No comments:

Post a Comment