Does anyone know of any work looking at artificial neural networks with fractal (but not necessarily self-similar) architectures or with several different learning rules/updates spread across many different spatiotemporal scales?
-
How exciting! Can’t wait to hear more about it
Actually, this paper from Tichko & Large (2019) used neural networks of neural oscillators to model infant rhythm perception. The learning rule was scaled by the oscillator's natural frequency, creating different dynamics of learning across space and time:https://nyaspubs.onlinelibrary.wiley.com/doi/full/10.1111/nyas.14050 …
-
-
Excellent! I’m rotating in a songbird lab right now, studying song learning and degradation in Zebra Finches, so this is very intriguing given the cultural differences in song amongst bird populations and the specificity with which juveniles learn their tutor’s song
-
Oh, yes! I think there are lots of parallels between the song-learning literature and cultural differences in music-learning.
Kraj razgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.