Content-type: text/html Downes.ca ~ Stephen's Web ~ Self-assembling neural networks can open new directions for AI research - TechTalks

Stephen Downes

Knowledge, Learning, Community

Artificial neural networks are a very limited copy of the real thing. " During the learning phase, they merely adjust their connection weights based on training examples. They lack the capacity to self-organize, grow, and adapt to new situations like biological neurons." This, suggests Ben Dickson in his summary of a paper called Towards Self-Assembling Neural Networks, is the next step in AI. "The vision outlined in the paper is to develop an AI system where neurons self-assemble, grow, and adapt based on the task at hand, emulating the natural processes observed in biological networks." This process is controlled by a "controlled by a second network operating within each neuron, termed the Neural Developmental Program (NDP)." The reality is that we're at the very beginning of the age of artificial intelligence.

Today: 0 Total: 1492 [Direct link] [Share]


Stephen Downes Stephen Downes, Casselman, Canada
stephen@downes.ca

Copyright 2024
Last Updated: May 22, 2024 4:24 p.m.

Canadian Flag Creative Commons License.

Force:yes