Tweetovi

Blokirali ste korisnika/cu @animesh_garg

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

  1. Prikvačeni tweet
    23. pro 2019.

    ML/AI and physical sciences Twitter - We are organizing a workshop on Differential Equations and Deep Learning at Maarten V. de Hoop -Rice Ron Fedkiw -Stanford Wilfrid Gangbo -UCLA -NTU Tom Goldstein -UMD Ricky T. Q. Chen -Toronto/Vector Gavin Portwood -LANL

    Poništi
  2. 31. sij

    It’s always great to visit MIT This time for a intro to robot learning lecture in deep learning short course Thanks to Alexandre Amini and for inviting me

    Poništi
  3. 31. sij
    Poništi
  4. 30. sij

    Good execution is often understated in the hype of chasing novelty Academic Tweeps, how do you teach this to students ?

    Poništi
  5. proslijedio/la je Tweet
    23. sij

    We released the code and data for GraspNet paper (). Code and data can be found at .

    Poništi
  6. 23. sij
    Poništi
  7. proslijedio/la je Tweet
    22. sij

    It is now an ICRA 2020 paper!! More details to come.

    Poništi
  8. proslijedio/la je Tweet
    20. sij

    We propose a method for in-hand manipulation that combines low-level manipulation primitives with a learned, mid-level policy that orchestrates these primitives. Our method keep the object firmly grasped while transporting it to distant goal poses.

    Prikaži ovu nit
    Poništi
  9. 19. sij

    Canadian winter and snow is harsh but there is always a silver lining. It makes for wonderful memes!

    Poništi
  10. 19. sij

    Very informative pair of blog posts. Questions to ponder -- "What is the B in BNNs doing without informative priors"

    Poništi
  11. proslijedio/la je Tweet
    13. sij

    An impressive conference put together by CNLS (& ). I’m excited to speak about my collaborative work in using neural odes for physics modeling with ! Video recordings of talks 2 be out soon

    Poništi
  12. 13. sij

    I'm always fascinated by robots in kitchen concepts or demos. So close yet so far! at it since 1968. 1968: Unimate in Kitchen ... 2015: Moley Robotics 2018: Sony Robot Kitchen Help me source other videos

    Poništi
  13. 11. sij

    I am deeply saddened by the fact that so many students from and other Canadian institutions were on the ill-fated flight. I am honored that the University has come out publicly in support of these tragic times.

    Poništi
  14. proslijedio/la je Tweet
    10. sij

    It is RSS season! As I read my student's paper drafts, I find myself giving the same comments, so here is my write-up of how to write an RSS paper: . Good luck everyone who is pushing for !

    Poništi
  15. proslijedio/la je Tweet
    8. sij

    is a new trend, and I found at the recently concluded . We compiled a list including our work with , and others including , that are relevant in this space! Take a look!

    Poništi
  16. proslijedio/la je Tweet
    Poništi
  17. 6. sij

    Excited to share thoughts about directions of my ongoing and future work on generalizable imitation in Robotics tomorrow at

    Poništi
  18. 30. pro 2019.

    Despite our attempt to create diversity in the community, this creates an insurmountable divide. I, for one, would not be admitted if this were the case. Publishing papers has a cost (compute, travel, equipment) that may not be available to all equitably.

    Prikaži ovu nit
    Poništi
  19. 30. pro 2019.

    What would be a workshop paper (good idea worth exploring, but utility/impact TBD) is now often a top-tier paper. And in doing so, as a community, we are devaluing the currency with which we evaluate our work, its impact, and allocate opportunities & resources.

    Prikaži ovu nit
    Poništi
  20. 30. pro 2019.

    In the quest for faster dissemination of ideas, every empirical trick is incentivized to be submitted as a paper - since PhD admission, fellowships, industry internships, talk invitations - all depend on it. 10 week UG internships often yield multiple papers. how is it possible?

    Prikaži ovu nit
    Poništi
  21. 30. pro 2019.

    And these 'papers' are what is used to hand out future opportunities and accolades. Perhaps the overall intellect of UG/MS researchers is still the same, but what we call research - in terms of novelty of idea, completeness of experiments, the potential utility - has gone down.

    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:

    ·