Roman Feiman @romanfeiman.bsky.social · 28/09/2026In those cases, an NN's learning trajectory is evidence of learnability, while the chick's lack of learning trajectory would be evidence against actual learning (i.e. evidence that X is innate). There's no paradox. 100
Jake Quilty-Dunn @quiltydunn.bsky.social · 28/09/2026the cases we cite in the paper can't be dismissed in this way precisely because the digital twins don't have the learning curve you assume they do 210
Roman Feiman @romanfeiman.bsky.social · 28/09/2026Sorry, I'm not following. I'm not assuming any particular learning curve or classic associationism, just differentiating between there being any evidence-dependent learning curve, or not. Would like to understand what you mean though! Maybe email is better? 200
Jake Quilty-Dunn @quiltydunn.bsky.social · 28/09/2026the assumption that NNs are classical associative learners and obey classical learning trajectories is one of the entrenched ideas we are trying to correct 010
