Unsupervised Thinking
a podcast about neuroscience, artificial intelligence and science more broadly

Contact us

            Subscribe to the Podcast via RSS
Subscribe to the podcast
or find us on iTunesStitcher, Spotify or YouTube

Questions, comments, critiques? Ideas for an episode?

You can put it in the comment box below.

Or you can email us at unsupervisedthinking.podcast@gmail.com.

You can also find us on Twitter @USTpodcast.

Finally, you can find Grace on Twitter @neurograce. Josh and Conor are Luddites without Twitter accounts. Our other frequent contributors are Yann, Nancy, and Alex.

10 comments:

  1. This comment has been removed by the author.

    ReplyDelete
  2. This comment has been removed by the author.

    ReplyDelete
  3. This comment has been removed by the author.

    ReplyDelete
  4. You guys! This is awesome. Keep this podcast going!

    ReplyDelete
  5. Suggestions: How a neuroscientist makes money? (Successful/Unsuccessful) neuroscience start-ups?

    ReplyDelete
  6. Could you share your podcast on Google Play: https://play.google.com/music/podcasts/publish?u=0#
    This makes it easier for Android users to find/listen to the podcast. Thanks.

    ReplyDelete
    Replies
    1. Thanks for the tip! I'll get to it when I'm back in the States in a few weeks (Google podcasts isn't available outside the US yet)

      Delete
  7. Hi, I'm a fan of your podcast.

    As an idea for an episode, I'm wondering if you all would be willing to discuss this new paper:
    https://www.nature.com/articles/s41586-019-1346-5

    This thread has some interesting discussion though it's a bit hard to follow and understand things:
    https://twitter.com/behrenstimb/status/1143947586556088320

    Some additional, related topics to discuss would be things like PCA, eigenspectra, efficient coding hypothesis, and power law distributions. Anyway this all stems from a selfish desire to understand these things myself so feel free to disregard lol.

    ReplyDelete
  8. Dear Grace,

    In episode 4 on Deep Learning you made the comment that lots of people feel that real intelligence would require emotions and consciousness is wrong. Well, if emotions aren't implemented, how are you going to handle CONTEXT correctly, Bayesian Statistics is hopelessly inadequate for correctly identifying and understanding the effects of context on interpretation. As for consciousness, how will you implement response timing as in Ready-Set-Go?

    ReplyDelete