The Transmitter

thetransmitter.bsky.social

21 October 2026 09:21

Neurons in models are often all the same—some researchers are trying to change that. #neuroskyence By @natmesanash.bsky.social www.thetransmitter.org/systems-neur...

Richard Gast

rgast.bsky.social

21 October 2026 09:20

It was a pleasure to discuss the various developments in modeling heterogeneous neural systems with @natmesanash.bsky.social, including my own work on the subject. She turned those conversations into an impressive article on a topic in computational neuroscience that continues to pick up speed!

Blake Richards

tyrellturing.bsky.social

21 October 2026 07:23

1/7) I fully agree with @kordinglab.bsky.social's response to @dlbarack.bsky.social. The motivation behind the "Strong NeuroAI" approach that myself and others have adopted is not simply an a priori philosophical position, but an empirical observation:

Blake Richards

tyrellturing.bsky.social

21 October 2026 07:22

5/7) I believe that the "Weak NeuroAI" program that @dlbarack.bsky.social advocates for (which I support), ends up being a form of the "Strong" program once we actually try to contend with *understanding* the neural networks that serve as a source of hypotheses and as a digital experimental testbed.

Blake Richards

tyrellturing.bsky.social

21 October 2026 07:22

4/7) It's only by tying the emergence of that manifold to the architecture, loss, learning rule, and *data/world* that the network was shaped by that we arrive at any additional knowledge gained by observing the manifold.