I would like to see human performance on the introduced benchmark, AmbiEnt, and have it added to Table 4.
HN user
almstimplmntd
2021-now: PhD student in machine learning + music data science
//
2017-2021: ML-engineer (Lighthouse A.I. --> Apple)
For AMZN software staff (especially in SF Bay Area): was it really 160K capped?
Obviously RSU grants / bonuses are excluded from this, but I still find that surprisingly low (as a salary cap) for FAANG, especially around SF.
Played with the gilbert gottfried speaker option, gave me serious Twin Peaks "Red Room" vibes.
A bit tangential but related: has anyone else noticed an increased frequency in the desired result of a Google search being the first one on the second page?
Particularly for things involving reviews or booking, I have noticed an uptick in the (subjectively) “right” result being kept off page 1... it seems almost like some post processing logic to favor specific sites/companies. The fact that the “right” result in these instances is almost always the first result makes me think the true Ranking function knows the result’s actual value.
As a (jazz) piano student myself, I am happy to see "To Master a Skill, Master Something Harder."
Pianist and YouTuber Nahre Sol has a great set of exercises based on modifying passages of Chopin etudes, which serve as excellent material to study both as a template and in their own right. [1][2] She varies harmony and rhythm, and takes the exercise through all 12-keys. Much harder than just learning the passage in the original key, but that's the whole point.
[1] https://www.youtube.com/watch?v=2xYfgJVm1Ns [2] https://flatfiv.com/products/piano-technique-intensive
So is “Quantum“ pure buzzword hype? Or is there a reasonable connection to the product?
This post is about Pyro (probabilistic programming framework, with a clever logo "pi rho"), not to be confused with https://github.com/irmen/Pyro4
Completely independent projects, just a name collision.