SuperMemo Algorithm - Revision history

SuperMemoHelp: /* Historical note: earlier releases of the algorithm */

2019-06-04T20:52:25Z

Historical note: earlier releases of the algorithm

← Older revision		Revision as of 16:52, 4 June 2019
Line 26:		Line 26:
	** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)		** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)
	*'''2015''' - [https://supermemo.guru/wiki/Algorithm_SM-17 Algorithm SM-17] is a successor to all prior versions of the algorithm in future SuperMemos. It had passed all important benchmarks, and was part of SuperMemo 17 in 2016		*'''2015''' - [https://supermemo.guru/wiki/Algorithm_SM-17 Algorithm SM-17] is a successor to all prior versions of the algorithm in future SuperMemos. It had passed all important benchmarks, and was part of SuperMemo 17 in 2016
	*'''2018''' - [https://supermemo.guru/wiki/Algorithm_SM-18 '''Algorithm SM-18'''] is the newest version of the spaced repetition algorithm introduced in SuperMemo 18. The new algorithm improves the way item difficulty is estimated		*'''2018''' - [https://supermemo.guru/wiki/Algorithm_SM-18 '''Algorithm SM-18'''] is the newest version of the spaced repetition algorithm introduced in SuperMemo 18. The new algorithm improves the way [https://supermemo.guru/wiki/Item_difficulty_in_Algorithm_SM-18 item difficulty is estimated]

	== See also ==		== See also ==

SuperMemoHelp at 12:37, 4 April 2019

2019-04-04T12:37:54Z

← Older revision		Revision as of 08:37, 4 April 2019
Line 30:		Line 30:
	== See also ==		== See also ==

	* [~~http~~://www.supermemopedia.com/wiki/Algorithm_SM-17_FAQs Algorithm SM-17 FAQs]		* [https://www.supermemopedia.com/wiki/Algorithm_SM-17_FAQs Algorithm SM-17 FAQs]
	* [http://super-memory.com/english/ol/sm2.htm Algorithm SM-2 (first algorithm for spaced repetitions on a computer)]		* [http://super-memory.com/english/ol/sm2.htm Algorithm SM-2 (first algorithm for spaced repetitions on a computer)]
	* [~~http~~://supermemopedia.com/wiki/Comparing_SuperMemo_with_other_applications_based_on_spaced_repetition Comparing SuperMemo with other applications based on spaced repetition]		* [https://supermemopedia.com/wiki/Comparing_SuperMemo_with_other_applications_based_on_spaced_repetition Comparing SuperMemo with other applications based on spaced repetition]
	* [~~http~~://supermemopedia.com/wiki/Spaced_repetition_algorithm_metric Spaced repetition algorithm metric]		* [https://supermemopedia.com/wiki/Spaced_repetition_algorithm_metric Spaced repetition algorithm metric]
	* [https://supermemo.guru/wiki/First_fast-converging_spaced_repetition_algorithm:_Algorithm_SM-5#Criticism_of_Algorithm_SM-5 Superiority of SuperMemo over Anki]		* [https://supermemo.guru/wiki/First_fast-converging_spaced_repetition_algorithm:_Algorithm_SM-5#Criticism_of_Algorithm_SM-5 Superiority of SuperMemo over Anki]
	* [~~http~~://supermemopedia.com/wiki/Open_letter_to_spaced_repetition_developers Open letter to spaced repetition developers]		* [https://supermemopedia.com/wiki/Open_letter_to_spaced_repetition_developers Open letter to spaced repetition developers]
	* [~~http~~://www.supermemopedia.com/index.php?title=Algorithm_SM-17&action=history Collaborative edits history of this file at SuperMemopedia]		* [https://www.supermemopedia.com/index.php?title=Algorithm_SM-17&action=history Collaborative edits history of this file at SuperMemopedia]

SuperMemoHelp at 07:32, 3 April 2019

2019-04-03T07:32:57Z

← Older revision		Revision as of 03:32, 3 April 2019
Line 25:		Line 25:
	** [[Glossary:U-Factor\|U-Factors]] now allow of correctly interpreting repetition delays of up to 15 years (previously 60 days only)		** [[Glossary:U-Factor\|U-Factors]] now allow of correctly interpreting repetition delays of up to 15 years (previously 60 days only)
	** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)		** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)
	*'''2015''' - Algorithm SM-17 is a successor to all prior versions of the algorithm in future SuperMemos. It had passed all important benchmarks, and was part of SuperMemo 17 in 2016		*'''2015''' - [https://supermemo.guru/wiki/Algorithm_SM-17 Algorithm SM-17] is a successor to all prior versions of the algorithm in future SuperMemos. It had passed all important benchmarks, and was part of SuperMemo 17 in 2016
	*'''2018''' - '''Algorithm SM-18''' is the newest version of the spaced repetition algorithm introduced in SuperMemo 18. The new algorithm improves the way item difficulty is estimated		*'''2018''' - [https://supermemo.guru/wiki/Algorithm_SM-18 '''Algorithm SM-18'''] is the newest version of the spaced repetition algorithm introduced in SuperMemo 18. The new algorithm improves the way item difficulty is estimated

	== See also ==		== See also ==

SuperMemoHelp: /* Historical note: earlier releases of the algorithm */

2019-03-06T19:19:32Z

Historical note: earlier releases of the algorithm

← Older revision		Revision as of 15:19, 6 March 2019
Line 25:		Line 25:
	** [[Glossary:U-Factor\|U-Factors]] now allow of correctly interpreting repetition delays of up to 15 years (previously 60 days only)		** [[Glossary:U-Factor\|U-Factors]] now allow of correctly interpreting repetition delays of up to 15 years (previously 60 days only)
	** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)		** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)
	*'''2015''' - Algorithm SM-17 is a successor to all prior versions of the algorithm in future SuperMemos. It ~~has~~ passed all important benchmarks, and was part of SuperMemo 17 in 2016		*'''2015''' - Algorithm SM-17 is a successor to all prior versions of the algorithm in future SuperMemos. It had passed all important benchmarks, and was part of SuperMemo 17 in 2016
	*'''2018''' - '''Algorithm SM-18''' is the newest version of the spaced repetition algorithm introduced in SuperMemo 18. The new algorithm improves the way item difficulty is estimated		*'''2018''' - '''Algorithm SM-18''' is the newest version of the spaced repetition algorithm introduced in SuperMemo 18. The new algorithm improves the way item difficulty is estimated

SMBeta: /* Historical note: earlier releases of the algorithm */

2019-03-06T17:16:57Z

Historical note: earlier releases of the algorithm

← Older revision		Revision as of 13:16, 6 March 2019
Line 22:		Line 22:
	**accounting for advanced repetitions by introducing O-Factor decrements derived from the concept of the ''spacing effect''		**accounting for advanced repetitions by introducing O-Factor decrements derived from the concept of the ''spacing effect''
	*'''2005''' - [http://www.freewarefiles.com/SuperMemo_program_13849.html SuperMemo 2004] introduced boundaries on A and B parameters computed from the '''Grade vs. Forgetting Index''' data. This plugs up a weakness in the algorithm that showed when importing repetitions from other applications (e.g. open source MemAid). If a large number of easy repetitions occurred at unnaturally long intervals (as after pre-training with another application), the graph might show reversed monotonicity that would temporarily affect the estimation of [[Glossary:A-Factor\|A-Factors]] (the speed of self-correction would be reversely proportional to the amount of flawed data). When boundaries are imposed, self-correction is instant, and the accuracy of [[Glossary:A-Factor\|A-Factor]] estimation increases with each repetition executed in SuperMemo		*'''2005''' - [http://www.freewarefiles.com/SuperMemo_program_13849.html SuperMemo 2004] introduced boundaries on A and B parameters computed from the '''Grade vs. Forgetting Index''' data. This plugs up a weakness in the algorithm that showed when importing repetitions from other applications (e.g. open source MemAid). If a large number of easy repetitions occurred at unnaturally long intervals (as after pre-training with another application), the graph might show reversed monotonicity that would temporarily affect the estimation of [[Glossary:A-Factor\|A-Factors]] (the speed of self-correction would be reversely proportional to the amount of flawed data). When boundaries are imposed, self-correction is instant, and the accuracy of [[Glossary:A-Factor\|A-Factor]] estimation increases with each repetition executed in SuperMemo
	*'''2011 '''- with [[Algorithm SM-15]], SuperMemo 15 eliminated two weaknesses of Algorithm SM-11 that would show up in heavily overloaded collections with very large item delays:		*'''2011 '''- with [https://supermemo.guru/wiki/Algorithm_SM-15 Algorithm SM-15], SuperMemo 15 eliminated two weaknesses of Algorithm SM-11 that would show up in heavily overloaded collections with very large item delays:
	** [[Glossary:U-Factor\|U-Factors]] now allow of correctly interpreting repetition delays of up to 15 years (previously 60 days only)		** [[Glossary:U-Factor\|U-Factors]] now allow of correctly interpreting repetition delays of up to 15 years (previously 60 days only)
	** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)		** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)

SMBeta: /* See also */

2019-03-06T17:16:10Z

SMBeta: /* See also */

2019-03-06T17:14:34Z

SMBeta: /* See also */

2019-03-06T17:13:36Z

SMBeta at 17:13, 6 March 2019

2019-03-06T17:13:14Z

← Older revision		Revision as of 13:13, 6 March 2019
Line 1:		Line 1:
			== Algorithm SM-18 ==

	SuperMemo had a fertile impact on the research of modeling long-term memory. One of the most interesting fruits of that research is the [http://www.super-memory.com/english/2vm.htm two component model of long-term memory]. Over the last two decades, it has been hinted on many occasions that the model may provide a sound theoretical basis for a better and more universal approach to [[Glossary:Spaced_repetition\|spaced repetition]]. Algorithm SM-18 is the most up-to-date implementation of the theory in a practical application. Due to its universal nature, the algorithm should, in the future, find its way to all SuperMemo products.		SuperMemo had a fertile impact on the research of modeling long-term memory. One of the most interesting fruits of that research is the [http://www.super-memory.com/english/2vm.htm two component model of long-term memory]. Over the last two decades, it has been hinted on many occasions that the model may provide a sound theoretical basis for a better and more universal approach to [[Glossary:Spaced_repetition\|spaced repetition]]. Algorithm SM-18 is the most up-to-date implementation of the theory in a practical application. Due to its universal nature, the algorithm should, in the future, find its way to all SuperMemo products.

	For details see: [https://supermemo.guru/wiki/Algorithm_SM-18 Algorithm SM-18]		For details see: [https://supermemo.guru/wiki/Algorithm_SM-18 Algorithm SM-18]

			== Historical note: earlier releases of the algorithm ==

			Although the presented algorithm may seem complex, you should find it easier to follow its steps once you understand the evolution of individual concepts such as [[Glossary:E-Factor\|E-Factor]], matrix of optimum intervals, optimum factors, [[Glossary:Forgetting curve\|forgetting curves]], mid-interval repetition, stability, retrievability, etc.

			*'''1985''' - Paper-and-pencil version of SuperMemo is formulated ([http://super-memory.com/english/ol/beginning.htm#Algorithm Algorithm SM-0]). Repetitions of whole pages of material proceed along a fixed table of intervals. See also: [http://super-memory.com/articles/paper.htm Using SuperMemo without a computer]
			*'''1987''' - [http://super-memory.com/articles/soft/sm2.htm First computer implementation] makes it possible to divide material into individual items. Items are classified into difficulty categories by means of [[Glossary:E-Factor\|E-Factor]]. Each difficulty category has its own optimum spacing of repetitions ([http://super-memory.com/english/ol/sm2.htm Algorithm SM-2])
			*'''1989''' - SuperMemo 4 was able to modify the function of optimum intervals depending on the student's performance ([http://super-memory.com/english/ol/sm4.htm Algorithm SM-4]). This was then the first algorithm in which the function of optimal intervals was adaptable
			*'''1989''' - SuperMemo 5 replaced the matrix of optimum intervals with the matrix of optimal factors (an optimum factor is the ratio between successive intervals). This approach accelerated the adaptation of the function of optimum intervals ([http://super-memory.com/english/ol/sm5.htm Algorithm SM-5])
			*'''1991''' - [http://super-memory.com/articles/soft/sm6.htm SuperMemo 6] derived optimal factors from [[Glossary:Forgetting curve\|forgetting curves]] plotted for each entry of the matrix of optimum factors. This could dramatically speed up the convergence of the function of optimum intervals to its ultimate value ([http://super-memory.com/english/ol/sm6.htm Algorithm SM-6]). This was then the first adaptable algorithm that would use regression to find the best fit to the actual memory performance data. Unlike SuperMemo 5, which could keep converging and diverging depending on the quality of the learning material and the learning process, SuperMemo 6 would get closer to the student's ultimate memory model with each day of learning
			*'''1995''' - [http://super-memory.com/articles/soft/sm8.htm SuperMemo 8] capitalized on data collected by users of [http://super-memory.com/articles/soft/sm6.htm SuperMemo 6] and [http://super-memory.com/articles/soft/sm7.htm SuperMemo 7] and added a number of improvements that strengthened the theoretical validity of the function of optimum intervals and made it possible to accelerate its adaptation, esp. in the early stages of learning ([http://super-memory.com/english/algsm8.htm Algorithm SM-8]). New concepts:
			**replacing [[Glossary:E-Factor\|E-Factors]] with absolute difficulty factors: [[Glossary:A-Factor\|A-Factors]]. Item difficulty was thus defined in terms of actual properties of human memory, and would not depend on the average difficulty of the learning material
			**fast approximation of [[Glossary:A-Factor\|A-Factors]] from the '''First Grade vs. A-Factor''' correlation graph and '''Grade vs. Forgetting index graph'''. This makes it possible to quickly guess item's difficulty before more data is available
			**real-time adjustment of the matrix of optimal factors based on the power approximation of the decline of optimum factors
			*'''2002''' - SuperMemo 2002 introduced the first SuperMemo algorithm that is resistant to interference from delay or advancement of repetitions. This makes it possible to safely delay repetitions ('''[[Postpone]]''') or advance repetitions ([[Subset learning\|'''Review''']]):
			**accounting for delayed repetitions by introducing the concept of repetition categories
			**accounting for advanced repetitions by introducing O-Factor decrements derived from the concept of the ''spacing effect''
			*'''2005''' - [http://www.freewarefiles.com/SuperMemo_program_13849.html SuperMemo 2004] introduced boundaries on A and B parameters computed from the '''Grade vs. Forgetting Index''' data. This plugs up a weakness in the algorithm that showed when importing repetitions from other applications (e.g. open source MemAid). If a large number of easy repetitions occurred at unnaturally long intervals (as after pre-training with another application), the graph might show reversed monotonicity that would temporarily affect the estimation of [[Glossary:A-Factor\|A-Factors]] (the speed of self-correction would be reversely proportional to the amount of flawed data). When boundaries are imposed, self-correction is instant, and the accuracy of [[Glossary:A-Factor\|A-Factor]] estimation increases with each repetition executed in SuperMemo
			*'''2011 '''- with [[Algorithm SM-15]], SuperMemo 15 eliminated two weaknesses of Algorithm SM-11 that would show up in heavily overloaded collections with very large item delays:
			** [[Glossary:U-Factor\|U-Factors]] now allow of correctly interpreting repetition delays of up to 15 years (previously 60 days only)
			** [[Glossary:Forgetting curve\|forgetting curves]] are now corrected for repetition delays beyond the maximum registered [[Glossary:U-Factor\|U-Factor]] value (preventing failed grades in delayed repetitions decreasing the estimates of the [[Glossary:Optimum interval\|optimum interval]] for standardly-scheduled items in the same category)
			*'''2015''' - Algorithm SM-17 is a successor to all prior versions of the algorithm in future SuperMemos. It has passed all important benchmarks, and was part of SuperMemo 17 in 2016
			*'''2018''' - '''Algorithm SM-18''' is the newest version of the spaced repetition algorithm introduced in SuperMemo 18. The new algorithm improves the way item difficulty is estimated

			== See also ==

			* [http://www.supermemopedia.com/wiki/Algorithm_SM-17_FAQs Algorithm SM-17 FAQs]
			* [[Algorithm SM-15\|SuperMemo Algorithm SM-15]]
			* [http://super-memory.com/english/ol/sm2.htm Algorithm SM-2]
			* [http://supermemopedia.com/wiki/Comparing_SuperMemo_with_other_applications_based_on_spaced_repetition Comparing SuperMemo with other applications based on spaced repetition]
			* [http://supermemopedia.com/wiki/Spaced_repetition_algorithm_metric Spaced repetition algorithm metric]
			* [http://supermemopedia.com/wiki/SuperMemo_or_Anki Comparison with algorithm used in Anki]
			* [http://supermemopedia.com/wiki/Open_letter_to_spaced_repetition_developers Open letter to spaced repetition developers]
			* [http://www.supermemopedia.com/index.php?title=Algorithm_SM-17&action=history Collaborative edits history of this file at SuperMemopedia]

SMBeta at 17:09, 6 March 2019

2019-03-06T17:09:44Z

← Older revision		Revision as of 13:09, 6 March 2019
Line 1:		Line 1:
	Go to [https://supermemo.guru/wiki/~~SuperMemo_Algorithm SuperMemo~~ Algorithm]		SuperMemo had a fertile impact on the research of modeling long-term memory. One of the most interesting fruits of that research is the [http://www.super-memory.com/english/2vm.htm two component model of long-term memory]. Over the last two decades, it has been hinted on many occasions that the model may provide a sound theoretical basis for a better and more universal approach to [[Glossary:Spaced_repetition\|spaced repetition]]. Algorithm SM-18 is the most up-to-date implementation of the theory in a practical application. Due to its universal nature, the algorithm should, in the future, find its way to all SuperMemo products.

			For details see: [https://supermemo.guru/wiki/Algorithm_SM-18 Algorithm SM-18]

← Older revision		Revision as of 13:16, 6 March 2019
Line 34:		Line 34:
	* [http://supermemopedia.com/wiki/Comparing_SuperMemo_with_other_applications_based_on_spaced_repetition Comparing SuperMemo with other applications based on spaced repetition]		* [http://supermemopedia.com/wiki/Comparing_SuperMemo_with_other_applications_based_on_spaced_repetition Comparing SuperMemo with other applications based on spaced repetition]
	* [http://supermemopedia.com/wiki/Spaced_repetition_algorithm_metric Spaced repetition algorithm metric]		* [http://supermemopedia.com/wiki/Spaced_repetition_algorithm_metric Spaced repetition algorithm metric]
	* [~~http~~://~~supermemopedia~~.~~com~~/wiki/~~SuperMemo_or_Anki Comparison with algorithm used in~~ Anki]		* [https://supermemo.guru/wiki/First_fast-converging_spaced_repetition_algorithm:_Algorithm_SM-5#Criticism_of_Algorithm_SM-5 Superiority of SuperMemo over Anki]
	* [http://supermemopedia.com/wiki/Open_letter_to_spaced_repetition_developers Open letter to spaced repetition developers]		* [http://supermemopedia.com/wiki/Open_letter_to_spaced_repetition_developers Open letter to spaced repetition developers]
	* [http://www.supermemopedia.com/index.php?title=Algorithm_SM-17&action=history Collaborative edits history of this file at SuperMemopedia]		* [http://www.supermemopedia.com/index.php?title=Algorithm_SM-17&action=history Collaborative edits history of this file at SuperMemopedia]

← Older revision		Revision as of 13:14, 6 March 2019
Line 31:		Line 31:

	* [http://www.supermemopedia.com/wiki/Algorithm_SM-17_FAQs Algorithm SM-17 FAQs]		* [http://www.supermemopedia.com/wiki/Algorithm_SM-17_FAQs Algorithm SM-17 FAQs]
	* [http://super-memory.com/english/ol/sm2.htm Algorithm SM-2]		* [http://super-memory.com/english/ol/sm2.htm Algorithm SM-2 (first algorithm for spaced repetitions on a computer)]
	* [http://supermemopedia.com/wiki/Comparing_SuperMemo_with_other_applications_based_on_spaced_repetition Comparing SuperMemo with other applications based on spaced repetition]		* [http://supermemopedia.com/wiki/Comparing_SuperMemo_with_other_applications_based_on_spaced_repetition Comparing SuperMemo with other applications based on spaced repetition]
	* [http://supermemopedia.com/wiki/Spaced_repetition_algorithm_metric Spaced repetition algorithm metric]		* [http://supermemopedia.com/wiki/Spaced_repetition_algorithm_metric Spaced repetition algorithm metric]