Peter Henderson

@PeterHndrsn

Machine learning, reinforcement learning, neuroscience, psych. Currently: Previously:

Stanford, CA
Vrijeme pridruživanja: veljača 2012.

Tweetovi

Blokirali ste korisnika/cu @PeterHndrsn

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

  1. Prikvačeni tweet
    27. stu 2017.

    Excited to show off our new investigative paper from joint MILA/McGill dialogue group: "Ethical Challenges in Data-Driven Dialogue Systems". Would love to hear feedback, so feel free to DM me! Big thanks to all co-authors!

    Prikaži ovu nit
    Poništi
  2. 15. sij

    Hey RL community: what's your fav continuous control task that's not MuJoCo locomotion? Looking to test out new algos, but am hoping to find something that's a bit more transferable to the real world than a one-legged hopper. Extra cool if related to climate change mitigation!

    Poništi
  3. proslijedio/la je Tweet
    4. sij

    NEWS: 1. All socio-economic scenarios take us over 2°C. Only one (using 'drastic geoengineering') gets us back down to 1.5°C. ~ New climate models, 2021 IPCC report 2. 'temperature rises of 3-4°C are likely now locked in' ~ Scientists at Pliocene Epoch conference, London 2019

    Prikaži ovu nit
    Poništi
  4. proslijedio/la je Tweet
    13. pro 2019.

    Are you at and trying to decide what workshop to go to today? Check out in West room 114+115. First retrospective by Leon Bottou just started!

    Prikaži ovu nit
    Poništi
  5. proslijedio/la je Tweet
    8. pro 2019.

    I will be speaking at the "Retrospectives" workshop: Talk title: "Active Symbols and Analogy-Making: Reflections on Hofstadter & Mitchell's Copycat project" I'll be at NeurIPS Wed.-Sat. & would be happy to chat with anyone there!

    Poništi
  6. proslijedio/la je Tweet

    Announcing our *AWESOME* group of invited speakers for our NeurIPS workshop (next week!!). Includes: Leon Bottou, , , , , , , and Percy Liang! 🤩🤩 See our schedule:

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    3. stu 2019.

    Excited to share new work!!! “Generalization through Memorization: Nearest Neighbor Language Models” We introduce kNN-LMs, which extend LMs with nearest neighbor search in embedding space, achieving a new state-of-the-art perplexity on Wikitext-103, without additional training!

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    16. lis 2019.

    Happy to share what , myself and have been working on over the last year: "Understanding the Curse of Horizon in Off-Policy Evaluation via Conditional Importance Sampling"

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet

    Urgent request to network: please like and retweet the tweet below as much as possible. is a dear friend that has made generational contributions to Canada and society at large. Sharing this is the least we can do to help. Cc:

    Prikaži ovu nit
    Poništi
  10. 1. lis 2019.

    Hi all, we decided to extend the deadline further until November 15th to accommodate all those who just made it through the previous month of deadlines. Considering sending something in to the AI Magazine Special Issue on Conversational AI (and spread the word)!

    Poništi
  11. proslijedio/la je Tweet

    We've decided to extend the workshop deadline one last time until after ICLR: Sunday, Sep 29!! This will be the last extension, since author notifications are required by Oct 1. Note that accepted papers will get guaranteed tickets! 👌

    Poništi
  12. proslijedio/la je Tweet
    17. ruj 2019.

    Scientists: the WH is coming to you to talk about securing research from foreign threats. If you value foreign-born researchers and want to make sure we strike the right balance and not jeopardize international collaboration, show up to these meetings!

    Poništi
  13. proslijedio/la je Tweet

    Announcement: we're extending the deadline for the Retrospectives workshop by 1 week, until Sunday Sep 22!! Take an hour to write up some thoughts about a past paper of yours (they don't need to be critical!) and submit by creating an issue on 🧐

    Poništi
  14. proslijedio/la je Tweet
    10. ruj 2019.

    Reminder: the deadline for the NeurIPS Retrospectives workshop is in less than a week (this Sunday)!! Also, it's not just for retrospectives: you can submit meta-analysis papers too! (analyze a set of papers that aren't your own) 🙂

    Poništi
  15. 6. ruj 2019.

    FYI, we've extended this to October 1st! Plenty of time after NeurIPS workshops to write a piece on conversational AI for the special issue!

    Poništi
  16. proslijedio/la je Tweet
    28. kol 2019.

    Update on ML Retrospectives: to make submitting easier (pull requests can be tricky), we're now encouraging users to submit through GitHub 'issues'. Just write your retrospective in our Markdown editor, and paste it into the issue template. There's also a preview option!! 🙏

    Prikaži ovu nit
    Poništi
  17. 28. kol 2019.

    Hey all! The deadline for submitting to the special AI Magazine issue on Conversational AI is September 10 (for now). We're looking for expository articles on conversational AI and welcome writing from a variety of topics/viewpoints within ! Site:

    Poništi
  18. 21. kol 2019.

    If you have any questions, send an email to the organizers (available on the ML Retrospectives website)! Current paper submission deadline for the workshop is September 15th. You can also submit retrospectives separately from the workshop on GitHub, see:

    Prikaži ovu nit
    Poništi
  19. 21. kol 2019.

    The "meta-analysis" track is a good venue for those who haven't yet published, but have takeaways from working closely in some area. You might have discovered a lot about your research topic that is valuable to the community, but you haven't found a good venue. This is that venue

    Prikaži ovu nit
    Poništi
  20. 21. kol 2019.

    Excited to be a part of the NeurIPS 2019 ML Retrospectives Workshop! We're accepting papers on two tracks: retrospectives (reflecting on your own work, see quoted) and "meta-analyses" (reflecting on a sub-field or collection of papers). The workshop CfP:

    Prikaži ovu nit
    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:

    ·