Spaced repetition systems are a well-known way to efficiently learn material. Recent innovations have applied machine learning to greatly improve their scheduling.

Interesting article about the FSRS algorithm

  • dragontamer@lemmy.world
    2·
    25 days ago

    Hmm. I was going to call bullshit in 10 words a day (20 new cards) but he claims 70% retention setting, which is much much lower than mine/default of 90%. I’ll start experimenting with some settings tonight…

    • emb@lemmy.worldM
      2·
      25 days ago

      10 new words also doesn’t necessarily mean 20 cards - not everyone (for every deck) goes for both recognition and production cards!

  • emb@lemmy.worldM
    1·
    25 days ago

    Good article! I knew FSRS was a big enhancement in Anki, but I didn’t know much else about how it came to be.

    One of the posts it links to, from the FSRS repo or creator, really lays out the advantage clearly:

    Aside from allowing the users to easily control their retention, FSRS has some other advantages when compared to Anki’s default algorithm. With FSRS, users have to do 20–30% fewer reviews than with Anki’s default algorithm to achieve the same retention level (2026-02-12 udpate: the figure is based on simulation results). FSRS is also much better at scheduling cards that have been reviewed with a delay, for example, if the user took a break from Anki for a few weeks or months.