Will Dabney

@wwdabney

Research scientist at DeepMind. On the critical path to AGI. Also, a persistent optimist.

London, England
Vrijeme pridruživanja: svibanj 2015.

Medijski sadržaj

  1. 9. lis 2019.

    Improved SPIRAL creating true art. The authors point out that constraining the generative agents leads to this type of visual abstraction. Maybe the remaining genius of humans is that we choose and experiment with our own constraints.

  2. 1. tra 2019.
    Odgovor korisniku/ci

    Here’s my point. I’ve plotted the effective flops for each GPU-based related work they cite against the training time. Their result (black dot) falls firmly in the trend. It’s certainly progress (compared to yellow dot nearby) but not crazy surprising.

  3. 5. velj 2019.

    Our new ICLR paper on LSTM training with distributed replay is camera ready! Great work from and on R2D2 with SOTA Atari-57 and DMLab-30:

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:

    ·