jarrett-ye

Good luck! I'm glad to hear that FSRS becomes more and more popular!

Lol, btw, my name is Jarrett Ye.

SM2 has less parameters, but there is not a systematic way to tweak them correctly. With the optimizer, it's easy to make FSRS adaptive to any learners. There is only one lever for user to tweak in FSRS: the desired retention.

Would you minding writing an intro for your app if it is using FSRS?

https://github.com/open-spaced-repetition/awesome-fsrs

Andy also tested this idea. You can read his notes here:

GPT-4 can probably estimate whether two flashcards are functionally equivalent

https://notes.andymatuschak.org/zJ7PMGzjcgBUoPjLUHBF9jn

GPT-4 can probably estimate whether one prompt will spoil retrieval of another

https://notes.andymatuschak.org/zK9Y15pCnRMLoxUahLCzdyc

Ebisu v2 doesn't perform well in our SRS Benchmark: https://github.com/open-spaced-repetition/srs-benchmark#:~:t...

An Experience Report from a Middle School Dropout.

Good feature! I guess someone could implement it via an add-on of Anki.

I guess what you need is Set Due Date, which allows you manually schedule cards in Anki.

Math Academy provided a self-service learning system with a novel spaced repetition algorithm which could take the hierarchical body of math into account.

Yes, we can improve FSRS in that way. But it requires to collect a lot of review data with cards' content, which breaks Anki's privacy policy. And FSRS is my side project which has squeezed all my free time... I'm not very motivated to improve it now.

Anki has a feature named Easy Days which allows you to spend less time on Anki on some days of the week.

Source: https://docs.ankiweb.net/deck-options.html?highlight=easy%20...

What does the 'Free' mean in the name?

The algorithm (FSRS) supports reviewing in advance or delay. It's free for users to decide the time of review. And it will adapt to the user's memory.

Meanwhile, spaced repetition is one essential technology to achieve free learning.

FSRS runs entirely locally and has no risk under others' control.

source: https://github.com/open-spaced-repetition/free-spaced-repeti...

To help more people understand FSRS and the R&D around it, I wrote this blog about my long history with FSRS and Anki.

Actually, we have found some algorithms which outperform FSRS[1]. Unfortunately, it's hard to deploy them in user's local device.

[1] https://github.com/open-spaced-repetition/srs-benchmark?tab=...

You don't need to beat FSRS with 700 millions of reviews. You can collect thousands of reviews for months and compare the ad-hoc algorithm with FSRS on those reviews.

Great! I added it into awesome-fsrs, a repository with implementations of FSRS in different programming languages and different software: https://github.com/open-spaced-repetition/awesome-fsrs

Would you consider to support FSRS in Mochi?

Here is an implementation of Leitner Box in Rust:

https://github.com/open-spaced-repetition/leitner-box-rs

- The algorithm is updated to FSRS-5.

- Forgetting curve is added into card info.

- Load balancing: within your fuzz range, Anki will now try to pick days that have fewer reviews waiting.

- Easy days: you can now tell Anki to try avoid certain days of the week.

- Simulator: you can get an estimate of your workload, either in reviews per day or in minutes of studying per day.

FSRS-based simulator could help user regulate the number of new cards they need to learn per day. The simulator will be integrated into Anki natively in the next release. But it's still an experimental feature. If the simulator is accurate enough, I plan to make it more automatically to support the idea that let the algorithm choose many new cards the user should have.

It's not the logic of FSRS. In fact, FSRS knows the probability of recalling any cards in your collection, which may means how urgently you need to see them. So it can be implemented via an add-on.

In the latest version, it only needs 64 review logs to full-optimize the model

You can set easy days via the FSRS4Anki Helper add-on: https://ankiweb.net/shared/info/759844606

And it also supports to postpone your backlog.

It's welcome to compare your algorithm with FSRS. Here is the benchmark of FSRS: https://github.com/open-spaced-repetition/fsrs-benchmark

I recommend the new algorithm of Anki: https://github.com/open-spaced-repetition/fsrs4anki

I also want to compare with SM-18, but I don’t know how to export the predictions of SM-18 from SuperMemo.

By the way ,SuperMemo CEO said that they will release the algorithm API. If it’s not very expensive, I will compare it with FSRS.

Yeah. I prefer to combine SRS and immersion learning, like the Refold method.

Yeah, here is the comparison between FSRS and SM-17: https://github.com/open-spaced-repetition/fsrs-vs-sm17

You can use a filtered deck and sort cards by relative overdueness in Anki.