Important Notice: Our web hosting provider recently started charging us for additional visits, which was unexpected. In response, we're seeking donations. Depending on the situation, we may explore different monetization options for our Community and Expert Contributors. It's crucial to provide more returns for their expertise and offer more Expert Validated Answers or AI Validated Answers. Learn more about our hosting issue here.

What is “fraction” in SuperMemo Algorithm SM-5?

0
Posted

What is “fraction” in SuperMemo Algorithm SM-5?

0

(#14257) (Joseph Michael Galero, Jul 03, 2008, 20:46:40) Question: I am trying to understand SM-5 and I have questions regarding Figure 3.4 (in http://www.supermemo.com/english/ol/sm5.htm): 1. fraction – described to be a number between 0 and 1. How is this determined? Answer: Whatever number you choose, as long as it is not too close to either 0 or 1, the algorithm will work. That value will determine stability vs. accuracy component of the optimization procedure. In other words, if the fraction is very high, you will get the OF matrix modified very fast, but its values may swing widely around the optimum. On the other hand, if you choose a very low value of the fraction, the algorithm will produce an accurate convergence of the OF matrix to its optimum/true value, however, the changes might be unacceptably slow, esp. that repetitions in SuperMemo may be spaced by years and you do not get that may repetitions in a lifetime to wait for the convergence! In practise, the fraction had to

Related Questions

What is your question?

*Sadly, we had to bring back ads too. Hopefully more targeted.

Experts123