• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
  • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI
    Reaction

    Which brain details matter in artificial neural networks

    A commenter argues that studying intelligence does not require artificial neural networks to replicate every detail of the brain.

    AN
    2 Sources, 4h ago, first seen 4h ago

    TLDR

    The commenter says choosing which biological details to leave out is important when using artificial neural networks to study intelligence. For work connecting the models with biomedical diseases, they say features such as neurotransmitters and brain topography will likely matter, and that researchers in NeuroAI are already adding such features.

    Combined views

    420

    2 Sources, first seen 4h ago

    Combined views

    420

    2 Sources, first seen 4h ago

    8 likes
    8 likes
    1 comments
    2 saves
    2 reposts
    1 comments
    2 saves
    2 reposts
    Featured Source

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    2 Sources

    @aran_nayebiIt's a common misconception that you need to emulate everything in the brain to get intelligent behavior. Rather, it depends on your question: if we want a *science* of intelligence (which is my goal!), then knowing what biological details to *discard* is actually quite critical. In this case, ANNs are operating at exactly the right level of representation we need for that. See here for more details: https://anayebi.github.io/files/NeuroAgents_LabPlanIntro_2024.pdf For connecting with biomedical diseases, then modeling things like neurotransmitters and topography will likely be important, and actually a fair number of folks in NeuroAI are actively adding these components into ANNs, even if they aren't directly relevant to intelligence or improving intelligent behavior.4h

    2 Sources

    @aran_nayebiIt's a common misconception that you need to emulate everything in the brain to get intelligent behavior. Rather, it depends on your question: if we want a *science* of intelligence (which is my goal!), then knowing what biological details to *discard* is actually quite critical. In this case, ANNs are operating at exactly the right level of representation we need for that. See here for more details: https://anayebi.github.io/files/NeuroAgents_LabPlanIntro_2024.pdf For connecting with biomedical diseases, then modeling things like neurotransmitters and topography will likely be important, and actually a fair number of folks in NeuroAI are actively adding these components into ANNs, even if they aren't directly relevant to intelligence or improving intelligent behavior.4h