ottaborra

5 ms·

31.
Ask HN: Any rising concerns about H5N1?
2 points by ottaborra 2y ago|3 comments
32.
by ottaborra 3y ago
It really feels like a scaling problem than anything else. It big companies go to shit as they do most of the time it's because tackling scale is a extremely difficult problem
33.
by ottaborra 3y ago
Man, I don't get it. Was I the only one put off the video game portion of the story? It seemed so out of place and enough for me to go "rubbish" and toss the series aside
34.
by ottaborra 3y ago
This sort of hubris has been the markings of the beginnings of the fall of companies. I am now fearful for Nvidia
35.
by ottaborra 3y ago
Why someone like Henri Poincare or Polya was not included in this list is beyond me. Both were interested in figuring how mathematicians did what they did and both have brilliant insights
36.
by ottaborra 3y ago
by taking steps to verify everything that was said
37.
by ottaborra 3y ago
Read the Butlerian Crusade
38.
by ottaborra 3y ago
The main point you're making is fair The only gripe I have is > Being an empirical science does not mean that the field is a "wild west" I think what you meant to say is: "Being an empirical science does not <b&gt
39.
by ottaborra 3y ago
This makes me wonder. Is deep learning as a field an empirical science purely because everyone is afraid of the math? It has the richness of modern day physics but for some reason most the practioners seem to want to keep thinking of it as
40.
by ottaborra 3y ago
My predictions: 1. state-space models make transformer based models obsolete 2. Cuda killer gets set loose by AMD 3. The world's first successful head transplant takes place 4. Children of Dune gets greenlit 5. Lex Friedman retires fro
41.
by ottaborra 3y ago
My predictions: 1. state-space models make transformer based models obsolete 2. Cuda killer gets set loose by AMD 3. The world's first successful head transplant takes place 4. Children of Dune gets greenlit 5. Lex Friedman retires fro
42.
by ottaborra 3y ago
did you mean kernel regression rather than kernel smoothing? I ask because: https://d2l.ai/chapter_attention-mechanisms-and-transformers... Quoting from a previous section > The attention mechanism allows us to aggregate
43.
by ottaborra 3y ago
Beautiful piece of math. Similar work: Traveling Words: A Geometric Interpretation of Transformers https://arxiv.org/abs/2309.07315