Research @ Neuromorphic

hexagons

There's a growing consensus among the brightest neuroscientists of our generation that criticality and nonlinear emergence is the key to neuromorphic activity. Pure localizationism is dead, and meta-network theory is only the beginning.

Meanwhile, existing artificial neural networks cannot achieve both scalability and biological fidelity: the result is an absurd expenditure of compute cycles to achieve mediocre results outside of any but the most impoverished domains of intelligence. A hummingbird with a brain the size of a grain of rice navigates from Maine to the Yucatan, calculates windsheer and multiple flight trajectories in milliseconds within a tiny energy budget, while we're uniformly impressed with a chatbot built atop an enormous bruteforce lookup table.

There is another path here: the next wave of cognitive science will be more attuned to biological models, both more humble and more deeply ambitious. It will be welcoming of the insights of anthropology, psychology, ethology, and sociology on the one hand, while more profoundly informed by synergetics and thermodynamics on the other.

Background

Neuromorphic is lead by T. Bartholomy, a cognitive scientist focused on the frontier between computational neuroscience, machine learning, and psychology.

Services

If this interests you, contact us: yo@bartholomy.ooo

< home