Vin Vashishta

@v_vashishta

Blunt talk about machine learning, biases, infosec.

Reno
Vrijeme pridruživanja: rujan 2010.

Tweetovi

Blokirali ste korisnika/cu @v_vashishta

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

  1. proslijedio/la je Tweet
    3. velj
    Poništi
  2. 2. velj

    I feel like I spend more time cleaning up and double checking my citations than I do writing the paper.

    Poništi
  3. proslijedio/la je Tweet
    30. sij

    DM 'ing multiple folk / convos at the same time is a whole level of danger and excitement I never knew existed

    Poništi
  4. proslijedio/la je Tweet
    28. sij

    New paper: Towards a Human-like Open-Domain Chatbot. Key takeaways: 1. "Perplexity is all a chatbot needs" ;) 2. We're getting closer to a high-quality chatbot that can chat about anything Paper: Blog:

    Prikaži ovu nit
    Poništi
  5. 28. sij
    Poništi
  6. 26. sij

    Retire the Purple. That’s his now.

    Poništi
  7. proslijedio/la je Tweet
    24. sij

    Q-LEARNING IN ENORMOUS ACTION SPACES VIA AMORTIZED APPROXIMATE MAXIMIZATION " We treat the search for the best action as another learning problem & replace the exact maximization over all actions with a maximization over a set of actions sampled from a ...

    Poništi
  8. proslijedio/la je Tweet
    23. sij

    Premature is the root of all evil. ~

    Poništi
  9. 22. sij

    Models that generate their own training data. This is a grail but there's more work to be done. Generative Teaching Networks: Accelerating Neural Architecture Search by Learning to Generate Synthetic Training Data

    Poništi
  10. 22. sij

    Followed by the mandatory, “That’s not a real earthquake. I’ve been in the ... quake”🤦‍♂️

    Poništi
  11. proslijedio/la je Tweet
    20. sij

    Hey twitter, I'm actively looking for work. Pretty much anything will do. Willing to relocate almost anywhere. I have IT experience and I am currently working on my Security+ Cert Any help would be much appreciated, just trying to feed my family. Please retweet for reach

    Poništi
  12. proslijedio/la je Tweet
    21. sij

    From blank canvas unfolds a scene: GAN-based model generates and modifies images based on continual linguistic instruction by Shikhar Sharma v/ Cc

    Poništi
  13. 19. sij

    Our whole approach to hiring is broken because the metrics guiding hiring decisions are based on 1 bad assumption. Busting the BS Metrics Behind Our Hiring Hot Mess

    Poništi
  14. 18. sij

    Go to the app on your iPad. Go to the news tab. Drop you iPad a short distance onto a couch or your lap. Enjoy animal pics and videos. Nice Easter Egg.

    Poništi
  15. 17. sij

    Tech companies...It’s time to go beyond basic annual training to career development with measurable impacts on retention and profits. Manufacturers to spend $26.2 billion on 'upskilling' in 2020 to attract and keep workers

    Poništi
  16. proslijedio/la je Tweet
    16. sij

    From blank canvas unfolds a scene: GAN-based model generates and modifies images based on continual linguistic instruction by Shikhar Sharma v/ Cc

    Poništi
  17. proslijedio/la je Tweet
    14. sij

    Necrotechnomancy: when your clients server is so far beyond end of life

    Poništi
  18. 13. sij

    The bio I’m writing for a conference in 5 months won’t be current when I take the stage to speak. That’s how fast our field is moving.

    Poništi
  19. 13. sij

    Be your own skeptic. PhDs to data scientists need to challenge assumptions based on simple metrics. Data science allow us to fill what averages leave to our imagination. A skeptical approach keeps assumptions out of our models.

    Poništi
  20. proslijedio/la je Tweet
    13. sij

    From blank canvas unfolds a scene: GAN-based model generates and modifies images based on continual linguistic instruction by Shikhar Sharma v/ Cc

    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:

    ·