Prithviraj Ammanabrolu

@rajammanabrolu

PhD Student at the Entertainment Intelligence Lab, . Formerly . NLP, RL, and interactive narrative.

Vrijeme pridruživanja: travanj 2019.

Tweetovi

Blokirali ste korisnika/cu @rajammanabrolu

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

  1. Prikvačeni tweet
    29. sij

    Ever wanted to walk around inside a fairy-tale? Well, now you can! Our new work brings stories alive - turning them into explorable, text-based interactive worlds.

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    3. velj

    I honestly can't tell if this is a real academic paper or an elaborate troll but either way, it's totally brilliant. "Randomly generated pseudo-profound bullshit titles increased the perceived profoundness of computer-generated abstract art" -

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    3. velj

    consider submitting to the 7th Workshop on Games and Natural Language Processing, to be held at check out for more info, and don't hesitate to contact me with any questions!

    Poništi
  4. proslijedio/la je Tweet
    31. sij

    My ICML reviews had better be good. I can’t afford a third dog.

    Poništi
  5. proslijedio/la je Tweet
    30. sij

    First game of neural network Interactive Fiction. I exit the Bank Vault full of red hair to find Archie "A bully, a heel, a battler, a devil, a lunatic, a lunatic with a heart of gold." Two lunatics, one man. Sounds about right.

    Poništi
  6. proslijedio/la je Tweet
    24. sij

    If you're doing work in AI and Creativity I highly recommend ICCC. Papers due March 1st!

    Poništi
  7. 29. sij

    We extract a knowledge graph from two genres of stories (mysteries and fairy-tales) and use it to build out a world. Extracted knowledge graphs and full games are tested by humans in terms of coherence, interestingness, and how well it matches the genre.

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

    For those interested in the topic: and I did and Then Raj did and with MSR But no one has beaten Zork yet

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    28. sij

    Hi, expert on AI story generation here 👋. Competent computer generated screenplays are more than 5 years away

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    28. sij

    What if we see level generation as a game, where each action changes the level in some way? Well, we could use RL to learn to "play" level generation! We introduce Procedural Content Generation via Reinforcement Learning (PCGRL), a new paradigm for PCG.

    Prikaži ovu nit
    Poništi
  11. 27. sij

    Now with updated expts+ArXiv+Code! We broadly show how knowledge graphs can be used to make RL-based language generation viable in combinatorially sized action spaces. ArXiv: Code:

    Poništi
  12. proslijedio/la je Tweet
    27. sij

    A lot has been said about the dangers of large-scale language model trained on the internet, like GPT-2. We fine-tuned a version of GPT-2 to avoid generating descriptions of non-normative behavior (killing, suicide, things inconsistent with social norms)

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

    Google Dataset Search is out of beta! 25 Million datasets indexed and growing quickly. Keep using it for your dataset needs!

    Poništi
  14. 22. sij

    The kick feels even better when it's delayed by half a year+

    Poništi
  15. proslijedio/la je Tweet

    Text-based games provide a platform to train RL agents that generate goal-driven language. Jericho framework by & provides benchmarks for scaling RL to combinatorially sized language action spaces:

    Poništi
  16. proslijedio/la je Tweet
    13. sij

    We are excited to announce our workshop 'Beyond "Tabula Rasa" in RL (BeTR-RL): Agents that remember, adapt, and generalize' w/ Ignasi Clavera, Kate Rakelly, , , CfP & more infos:

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    1. sij

    ░░░░░░░░░░░░░░░ 0%

    Poništi
  18. proslijedio/la je Tweet
    29. pro 2019.

    “To start a PhD in ML, without insider referral, you need to do work equivalent to half of a PhD. Hence, in Apr 2019, I decided to dedicate all my time until Jan 2020 to publish in either NeurIPS or ICLR. If I fail, I would become a JavaScript programmer.” — ‼️

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet

    What is hidden in the structure of Choose-Your-Own-Adventure novels?

    Prikaži ovu nit
    Poništi
  20. Featuring yours truly! Finally a thing where my full name is an advantage

    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:

    ·