Hannah Sheahan

@hannahsheahan

Planning to learn something about learning and planning

Vrijeme pridruživanja: kolovoz 2013.

Tweetovi

Blokirali ste korisnika/cu @hannahsheahan

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @hannahsheahan

  1. proslijedio/la je Tweet
    29. sij

    Today I will be teaching my undergrad course "How to build a brain from scratch" for the 2nd year running. I've put the materials online - include a document with all lecture slides and notes, which is about as long as a decent novel. Enjoy!

    Prikaži ovu nit
    Poništi
  2. 9. pro 2019.

    Andrew will be a brilliant PI - this is a great opportunity for grad students or anyone looking for a job here in Oxford!

    Poništi
  3. proslijedio/la je Tweet
    26. srp 2019.

    I'm recruiting graduate students to start Fall 2020 in my new lab at Psychology! Come research motor & reinforcement learning using behavioral experiments, neuropsychology, fMRI, and modeling! Visit for some more info Please RT! :)

    Poništi
  4. proslijedio/la je Tweet
    15. srp 2019.

    "A theory of structure learning in the parietal cortex". New preprint from

    Poništi
  5. proslijedio/la je Tweet
    18. svi 2019.

    People say the brain is Bayesian. But does it really understand Bayesian principles or did it just find a Bayesian stats package and decide to try it out?

    Poništi
  6. proslijedio/la je Tweet
    16. svi 2019.

    People were talking about when they first wrote code, and were throwing around numbers like 6 or 9. I was 30. It’s not magic. You do not need to apprentice to a wizard when you are 11.

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    16. tra 2019.

    Where does value come from? In our new preprint () we explore some of the open challenges of Reinforcement Learning and reward-based decision-making more broadly. THREAD 1/13

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    29. sij 2019.

    New review discussing current models of how brain networks could implement the back-propagation algorithm! Work with Rafal Bogacz. We discuss models from Yoshua Bengio, Walter Senn and others :)

    Poništi
  9. proslijedio/la je Tweet
    8. tra 2019.

    barcelona summer school for advanced modelling of behaviour, 4th to 10th september 2019 - structured talks, hands on tutorials, group projects. aimed at PhDs and postdocs in cognitive and computational neuroscience. apply now!

    Poništi
  10. proslijedio/la je Tweet
    11. ožu 2019.

    Check out our new paper 'Neural structure mapping in human probabilistic reward learning' in with and Bernhard Spitzer: tl;dr: a thread!

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    23. sij 2019.

    The advice I continue giving folks looking into PhD programs is to look for an advisor who is kind. Don’t look for the “most brilliant” advisor. Look for the person who is kind, who mentors you, lifts you up. And pro-tip: kind folks are usually the most brilliant.

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    30. srp 2018.

    A real robot hand, trained with same learning algorithm and code as OpenAI Five, has learned human-like motions to rotate objects:

    Poništi
  13. proslijedio/la je Tweet
    11. srp 2018.

    Tomorrow Felix Hill will present our work: Measuring abstract reasoning in neural networks. Paper: Blog:

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    10. srp 2018.

    What do 🐴🐴🐴 have in common with 1️⃣2️⃣3️⃣ ... ? Check out my new with about how newly learned values neurally align with the mental number line

    Poništi
  15. proslijedio/la je Tweet
    26. lip 2018.

    The Cognition and Action Lab is recruiting a new . Flexible in terms of research focus and methodology—we’re interested in all aspects of human performance. More info at . Please RT and share !

    Poništi
  16. proslijedio/la je Tweet

    Most of you know me as a successful neuroscientist / deep learning researcher but I have a story that I want to share briefly. I grew up in a group home, which is basically an orphanage.

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet

    Don’t worry if your results aren’t significant. It’s likely that the phenomenon you’ve dedicated your life towards studying is so noisy and hard to measure that nothing of subtance can be said about it at all. It’s not your fault, it’s just hopeless, that’s all!

    Poništi
  18. 23. svi 2018.

    Great to see and Amar finally launching ! Brilliant - I'm excited to see what results from this dream team

    Poništi
  19. proslijedio/la je Tweet

    Will be hiring a postdoc in movement science. Video based pose tracking is getting good. Lets do movement science on >1000 subjects.

    Poništi
  20. proslijedio/la je Tweet
    2. svi 2018.

    If you want to know how probabilistic inference can be tied to optimal control, I just put up a new tutorial on control as inference: This expands on the control as inference lecture in my class:

    Poništi

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.

    Možda bi vam se svidjelo i ovo:

    ·