Hi,
I found this on Anki's website ( http://ichi2.net/anki/wiki/FrequentlyAskedQuestions ). The author explains the algorith that he used in Anki:
"Anki's algorithm is based on SM2, but differs from it in some respects.
* SM2 defines an initial interval of 1 day then 6 days. With Anki, you have the choice of 1, 3-5 and 7-9 days depending on your initial answer. After that, the intervals are increased in largely the same way. Anki understands that it can be necessary to see a new card a number of times before you're able to memorize it, and those initial 'failures' don't mean you need to be 'punished' by being shown the failed card many times over the course of a few days. Performance during the learning stage does not reflect performance in the retaining stage.
* Anki uses 4 choices for answering, not 6. There is only one 'fail' choice, not 3. The reason for this is that failure comprises a small amount of total reviews, and thus adjusting a card's ease can be sufficiently done by simply varying the positive answers.
* Answering cards later than scheduled will be factored into the next interval calculation, so you receive a boost to cards that you were late in answering but still remembered.
* SM2 does not define a specific order to show the cards in. Anki has support for different priorities, and the user can choose the order that cards due for review are displayed in. The default order shows oldest cards first, which makes the first half of a review session easier and reduces the chances of a user falling behind in their reviews and becoming demotivated.
* Like SM2, Anki's failure button resets the card interval by default. But the user can choose to make the card take some steps back rather than being reset completely. Also, you can elect to review failed mature cards on a different day, instead of the same day.
* Learning new material is integrated into the review process. Failing cards results in them being presented again for review within 10 minutes by default, instead of all at the end like SuperMemo's final drill option. There is also an option to keep failed cards to the end.
* 'Remembered easily' not only increments the ease factor, but adds an extra bonus to the current interval calculation. Thus, answering 'remembered easily' is a little more aggressive than the standard SM2 algorithm.
* Successive failures while cards are 'young' (under 21 days) do not result in further decreases to the card's ease. A common complaint with the standard SM algorithms is that repeated failings of a card cause the card to get stuck in "low interval hell". In Anki, the initial acquisition process does not influence a card's ease as much as it does in SuperMemo.
* Ease factors start off at 2.5 for the very first new card, but subsequent new cards are based on the average ease factor of the deck. Thus in a deck where the average easiness is very low, new cards will start out with more conservative intervals. Note that once a new card has been answered once, future scheduling is independent of other cards and thus its progress is predictable, unlike SM3+."
I'm a user of Supermemo, and I was wondering what to think about it. What do you think about SM's repetition algorithm. I'm especially interested in this bit:
"A common complaint with the standard SM algorithms is that repeated failings of a card cause the card to get stuck in "low interval hell". In Anki, the initial acquisition process does not influence a card's ease as much as it does in SuperMemo."
Since I've been using SM for about 1,5 year now, I seem to have noticed something similar to what is described here with respect to a few items.
What do you think about it? And if this is true, then is there any way to sort it out?
regards
I found this on Anki's website ( http://ichi2.net/anki/wiki/FrequentlyAskedQuestions ). The author explains the algorith that he used in Anki:
"Anki's algorithm is based on SM2, but differs from it in some respects.
* SM2 defines an initial interval of 1 day then 6 days. With Anki, you have the choice of 1, 3-5 and 7-9 days depending on your initial answer. After that, the intervals are increased in largely the same way. Anki understands that it can be necessary to see a new card a number of times before you're able to memorize it, and those initial 'failures' don't mean you need to be 'punished' by being shown the failed card many times over the course of a few days. Performance during the learning stage does not reflect performance in the retaining stage.
* Anki uses 4 choices for answering, not 6. There is only one 'fail' choice, not 3. The reason for this is that failure comprises a small amount of total reviews, and thus adjusting a card's ease can be sufficiently done by simply varying the positive answers.
* Answering cards later than scheduled will be factored into the next interval calculation, so you receive a boost to cards that you were late in answering but still remembered.
* SM2 does not define a specific order to show the cards in. Anki has support for different priorities, and the user can choose the order that cards due for review are displayed in. The default order shows oldest cards first, which makes the first half of a review session easier and reduces the chances of a user falling behind in their reviews and becoming demotivated.
* Like SM2, Anki's failure button resets the card interval by default. But the user can choose to make the card take some steps back rather than being reset completely. Also, you can elect to review failed mature cards on a different day, instead of the same day.
* Learning new material is integrated into the review process. Failing cards results in them being presented again for review within 10 minutes by default, instead of all at the end like SuperMemo's final drill option. There is also an option to keep failed cards to the end.
* 'Remembered easily' not only increments the ease factor, but adds an extra bonus to the current interval calculation. Thus, answering 'remembered easily' is a little more aggressive than the standard SM2 algorithm.
* Successive failures while cards are 'young' (under 21 days) do not result in further decreases to the card's ease. A common complaint with the standard SM algorithms is that repeated failings of a card cause the card to get stuck in "low interval hell". In Anki, the initial acquisition process does not influence a card's ease as much as it does in SuperMemo.
* Ease factors start off at 2.5 for the very first new card, but subsequent new cards are based on the average ease factor of the deck. Thus in a deck where the average easiness is very low, new cards will start out with more conservative intervals. Note that once a new card has been answered once, future scheduling is independent of other cards and thus its progress is predictable, unlike SM3+."
I'm a user of Supermemo, and I was wondering what to think about it. What do you think about SM's repetition algorithm. I'm especially interested in this bit:
"A common complaint with the standard SM algorithms is that repeated failings of a card cause the card to get stuck in "low interval hell". In Anki, the initial acquisition process does not influence a card's ease as much as it does in SuperMemo."
Since I've been using SM for about 1,5 year now, I seem to have noticed something similar to what is described here with respect to a few items.
What do you think about it? And if this is true, then is there any way to sort it out?
regards