Fabled
Fabled on Hacker News
Coincidentally I was watching reviews of some games from AlphaZero vs. Stockfish 9 and people were amazed at the "human-like" nature of some of AlphaZero's decisions. It was seemingly sacrificing material with no obvious gain, only to come back much later in the game with a stronger strategic position. While Stockfish evaluates all possible moves it is limited by computation time and can't see too far into the future. AlphaZero responded much faster because it has already precomputed the expected probability of winning for most branches, so it can seemingly blunder when in fact it is following the path that led to victory most often.
Posted by
Fabled on Sunday December 22, 2019 @08:27AM
from the media.mit.edu dept.
I don't think it's always best to stick to it blindly. You start shifting your perspective from "I want to achieve my goal" to "I don't want to break the chain" and if you falter once you feel weak and worthless, and quit completely altogether. You just have to constantly remind yourself why you are doing this, know that repetition is the key and it's OK to miss some days here and there.
Give him a break, he's still recovering ;)
Haha, I meant the actual coding part. I'm sure the problem solving would take me longer, it's just that once I have a solution in mind I can code it in 10 minutes in Python, since I'm very familiar with it and use it almost on a daily basis.
Having to think about basic things such as how to instantiate hash maps, what is the syntax for control flow, does this language have this particular functionality. This can take hours spent in reading documentation and debugging.
From what I've seen from last year, the problems arent't that hard (they're meant to be solved within a day), so the real time consuming part is the coding.
I'm thinking on doing them in Rust and Clojure. I've dabbled in Clojure before but nothing serious and I'd like to build up some more familiarity. Rust because it's the first systems programming language that grabbed my attention; plus it's always good to have a varied selection of tools in your proverbial toolbox.
I know that I can probably do them in 10 minutes in Python, but that's not going to be as fun or valuable, especially since I'm not going to be able to compete for the leaderboard due to timezone/work.
Usually on the "Sound of.." playlist there is a link to the Pulse and Edge in the description. Otherwise the genre might be too obscure and a playlist for it not generated. Also, the "Sound of" playlists are stored under the "Sounds of Spotify" account, while the pulse and edge under "Particle Detector", so you might try searching directly in that profile.
In case you didn't know, Spotify categorizes music into genres behind the scenes. You can use this site to find out what genres your favorite artists are categorized as:
http://everynoise.com/engenremap.htmlYou can use this information to then check out Spotify's auto-generated playlists for each genre. They have at least three types for each one: "The Sound of <genre>", containing definitive representation of the genre, "The Pulse of <genre>", containing songs that fans of the genre listen to now, and "The Edge of <genre>", with unpopular songs (not necessarily of the same genre) that fans of the genre listen to.
This has been a great way for me to find new music that I like, especially "The Pulse". I even created a small script that takes a spotify playlist as input, parses all the artists, converts to genres, and creates a new playlist with "recommendations" based on the Pulse playlists, with each genre represented based on its percentage in the initial playlist.
Basically, I never use Discover Weekly, because I know it will eventually converge just like all my Pandora stations that cycle through the same 20 songs after a few months.