DOCSLIB.ORG

A Queueing Network Model for Spaced Repetition Siddharth Reddy, Igor Labutov, Siddhartha Banerjee Sgr45, Iil4, Sbanerjee @Cornell.Edu { } Introduction Experiments

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
A Queueing Network Model for Spaced Repetition Siddharth Reddy, Igor Labutov, Siddhartha Banerjee Sgr45, Iil4, Sbanerjee @Cornell.Edu { } Introduction Experiments A Queueing Network Model for Spaced Repetition Siddharth Reddy, Igor Labutov, Siddhartha Banerjee sgr45, iil4, sbanerjee @cornell.edu f g Introduction Experiments The ability to retain a large number of new ideas in memory is an essential component of human 0.9 30000 learning. In recent times, there has been a growing body of work that attempts to `engineer' this process { creating tools that enhance the learning process by building on the scientific understanding 0.8 25000 of human memory. Flashcards are one such tool that use the idea of spaced repetition to overcome the human `forgetting curve'. Though they have been around for a while in the physical form, a 0.7 new generation of spaced repetition software such as SuperMemo, Anki, and Mnemosyne allow a 20000 0.6 much greater degree of control and monitoring of the process. As these software applications grow popular, there is a need for formal models for reasoning about and optimizing their operations. In 0.5 15000 this work, we use ideas from queueing theory to develop such a formal model for one of the simplest and oldest spaced repetition systems: the Leitner system. 0.4 Validation AUC 10000 0.3 Key Contributions 5000 0.2 Frequency (number of interactions) Formalize the Leitner spaced repetition system using ideas from queueing theory • 0.1 0 Validate the exponential forgetting curve on large-scale log data from the Mnemosyne flashcard 1 2 3 4 5 6 7 8 9 10 • software Deck (qij) Model 3: φ(θ β ) j − i Verify key predictions of our queueing model using a controlled experiment on Amazon Mechanical Model 5: exp ( θ d /n ) Model 10: exp ( θ d /n ) • − · ij ij − i · ij ij Turk Model 7: exp ( θ/n ) Model 12: exp ( θ /n ) − ij − i ij Model 8: exp ( θ d /q ) Model 13: exp ( θ d /q ) − · ij ij − i · ij ij Model 9: exp ( θ/q ) Model 14: exp ( θ /q ) Memory Model − ij − i ij We use a variant on the exponential forgetting curve to model the decay of human memory over 0.025 time. The probability of a student recalling an item is as follows. Maximum Arrival Rate λext∗ Empirical D P[recall] = exp θ 0.020 Simulated (Clocked Delay) − · s Simulated (Expected Delay) where θ is the item difficulty, D is time elapsed since previous review, and s is memory strength. 0.015 Leitner Queue Network (Items Per Second) n λ 0.010 mastered items! 0.005 Throughput 0.000 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 Arrival Rate λext (Items Per Second) 1! 2! 3! 4! 5! 70 Deck 1 60 Deck 2 Deck 3 Deck 4 50 Deck 5 Mastered 40 new items! 30 Number of Items 20 New items arrive into deck 1 according to a Poisson process with rate λext. The routing probability matrix is P , where 10 8 < P[recall s = i; ] if i < n j = i + 1 P = 1 [recallj s =· i; ] if (i > 1^ j = i 1) i = j = 1 0 ij P 0.00 0.02 0.04 0.06 0.08 0.10 0.12 : 0 − j · otherwise^ − _ Arrival Rate λext (Items Per Second) Items exit the system from deck n with probability P[recall s = n; ]. The service rate for deck i is indicated by µ , and the user's work rate budget (e.g., thej maximum· number of items the user i 1.0 can review per day) is given by U. We are interested in finding µi that maximize the steady-state λ = 0.004 (Before Phase Transition) n ext P throughput of the system such that the budget constraint µi U is satisfied. Formally, we λext = 0.015 (Before Phase Transition) ≤ i=1 0.8 λext = 0.048 (After Phase Transition) must solve the following static planning problem. λext = 0.114 (After Phase Transition) maximize λext n 0.6 X subject to µi U ≤ i=1 λext + P11λ1 + P21λ2 = λ1 0.4 P12λ1 + P32λ3 = λ2 . 0.2 Pn 1;nλn 1 + Pnnλn = λn − − µi; λi 0 i Fraction of Items Seen During Session ≥ 8 λi < µi i 8 0.0 The algorithm for selecting the next item to present to the user is simple: sample deck i with 1 2 3 4 5 6 µi probability Pn , and select the item at the top of the sampled deck. Deck i=1 µi Full paper and code available at http://siddharth.io/leitnerq.
Load more

Recommended publications
  • A Queueing-Theoretic Foundation for Optimal Spaced Repetition
    A Queueing-Theoretic Foundation for Optimal Spaced Repetition Siddharth Reddy [email protected] Department of Computer Science, Cornell University, Ithaca, NY 14850 Igor Labutov [email protected] Department of Electrical and Computer Engineering, Cornell University, Ithaca, NY 14850 Siddhartha Banerjee [email protected] School of Operations Research and Information Engineering, Cornell University, Ithaca, NY 14850 Thorsten Joachims [email protected] Department of Computer Science, Cornell University, Ithaca, NY 14850 1. Extended Abstract way back to 1885 and the pioneering work of Ebbinghaus (Ebbinghaus, 1913), identify two critical variables that de- In the study of human learning, there is broad evidence that termine the probability of recalling an item: reinforcement, our ability to retain a piece of information improves with i.e., repeated exposure to the item, and delay, i.e., time repeated exposure, and that it decays with delay since the since the item was last reviewed. Accordingly, scientists last exposure. This plays a crucial role in the design of ed- have long been proponents of the spacing effect for learn- ucational software, leading to a trade-off between teaching ing: the phenomenon in which periodic, spaced review of new material and reviewing what has already been taught. content improves long-term retention. A common way to balance this trade-off is spaced repe- tition, which uses periodic review of content to improve A significant development in recent years has been a grow- long-term retention. Though spaced repetition is widely ing body of work that attempts to ‘engineer’ the process used in practice, e.g., in electronic flashcard software, there of human learning, creating tools that enhance the learning is little formal understanding of the design of these sys- process by building on the scientific understanding of hu- tems.
    [Show full text]
  • The Spaced Interval Repetition Technique
    The Spaced Interval Repetition Technique What is it? The Spaced Interval Repetition (SIR) technique is a memorization technique largely developed in the 1960’s and is a phenomenal way of learning information very efficiently that almost nobody knows about. It is based on the landmark research in memory conducted by famous psychologist, Hermann Ebbinghaus in the late 19th century. Ebbinghaus discovered that when we learn new information, we actually forget it very quickly (within a matter of minutes and hours) and the more time that passes, the more likely we are to forget it (unless it is presented to us again). This is why “cramming” for the test does not allow you to learn/remember everything for the test (and why you don’t remember any of it a week later). SIR software, notably first developed by Piotr A. Woźniak is available to help you learn information like a superhero and even retain it for the rest of your life. The SIR technique can be applied to any kind of learning, but arguably works best when learning discrete pieces of information like dates, definitions, vocabulary, formulas, etc. Why does it work? Memory is a fickle creature. For most people, to really learn something, we need to rehearse it several times; this is how information gets from short‐term memory to long‐ term memory. Research tells us that the best time to rehearse something (like those formulas for your statistics class) is right before you are about to forget it. Obviously, it is very difficult for us to know when we are about to forget an important piece of information.
    [Show full text]
  • Learning Efficiency Correlates of Using Supermemo with Specially Crafted Flashcards in Medical Scholarship
    Learning efficiency correlates of using SuperMemo with specially crafted Flashcards in medical scholarship. Authors: Jacopo Michettoni, Alexis Pujo, Daniel Nadolny, Raj Thimmiah. Abstract Computer-assisted learning has been growing in popularity in higher education and in the research literature. A subset of these novel approaches to learning claim that predictive algorithms called Spaced Repetition can significantly improve retention rates of studied knowledge while minimizing the time investment required for learning. SuperMemo is a brand of commercial software and the editor of the SuperMemo spaced repetition algorithm. Medical scholarship is well known for requiring students to acquire large amounts of information in a short span of time. Anatomy, in particular, relies heavily on rote memorization. Using the SuperMemo web platform1 we are creating a non-randomized trial, inviting medical students completing an anatomy course to take part. Usage of SuperMemo as well as a performance test will be measured and compared with a concurrent control group who will not be provided with the SuperMemo Software. Hypotheses A) Increased average grade for memorization-intensive examinations If spaced repetition positively affects average retrievability and stability of memory over the term of one to four months, then consistent2 users should obtain better grades than their peers on memorization-intensive examination material. B) Grades increase with consistency There is a negative relationship between variability of daily usage of SRS and grades. 1 https://www.supermemo.com/ 2 Defined in Criteria for inclusion: SuperMemo group. C) Increased stability of memory in the long-term If spaced repetition positively affects knowledge stability, consistent users should have more durable recall even after reviews of learned material have ceased.
    [Show full text]
  • Arichardson-Klavehn Rbjork 2002
    1096 Memory, Development of Schneider W and Bjorklund DF (1998) Memory. In: Kuhn Schneider W and Pressley M (1997) Memory Development D and Siegler RS (eds) Handbook of Child Psychology, between 2 and 20, 2nd edn. Mahwah, NJ: Lawrence vol. 2, Cognition, Perception, and Language, 5th edn, Erlbaum. pp. 467±521. New York, NY: John Wiley. Memory, Long-term Introductory article Alan Richardson-Klavehn, Goldsmiths College, University of London, London, UK Robert ABjork, University of California, Los Angeles, USA CONTENTS Definition and classification of long-term memory The constructive character of long-term memory The dynamic character of long-term memory Conclusion Long-term memory is central to cognitive function- interconnected and cannot be understood in ing. Taking a wide variety of forms, from skills to isolation from each other. (See Information Pro- general knowledge to memory for personal experi- cessing) ences, it is characterized by dynamic interactions between encoding and retrieval processes and by constructive processes, and thus differs fundamen- Distinguishing between Short-term and tally from current human-made information storage Long-term Memory systems. In everyday discourse, long-term memory is usu- ally distinguished from short-term memory in DEFINITION AND CLASSIFICATION OF terms of the time that has elapsed since information LONG-TERM MEMORY was encoded. Moreover, it is not unusual to find memory that persists over days or weeks being The ability to retain information over long periods described as short-term memory. In psychology, is fundamental to intelligent thought and behavior. however, the terms long-term and short-term Memory is the `glue', in effect, that holds our intel- memory have come to have specialized meanings lectual processes together, from perception, atten- that stem from a distinction made by William tion, and language, to reasoning, decision-making, James in 1890.
    [Show full text]
  • Useful Links for Language Learners
    Useful Links for Language Learners All these links are for free resources that can help language learners. Some of these resources have paid options, but you do not have to sign up for the paid option to utilize their resources and learn a new language! Duolingo: This is a language teaching app that uses a combination of spaced repetition, game style learning and motivation, stories, and tips to share grammar, pronunciation, and cultural differences related to the language. Do not forget to explore the site beyond the lessons or you may miss out on some of what Duolingo has to offer. Forvo: This is a library of language audio clips. That is to say, native speakers say words in their language and share them at Forvo. Others can listen to those clips. If you create an account, you can also download the clips for non-commercial use (i.e. studying). If you are making your own digital flashcards, this can be a very useful site to visit. Google Images: This is like regular Google, but it searches primarily for images. This is useful for language learners, because doing an image search for a word in a foreign language can visually show how that word is perceived. It may show you an unexpected meaning or nuance that a simple translation cannot share; for instance, the Italian word ‘ragazzo’ could be translated into English as ‘boy’, but doing an image search will mostly reveal teenage boys, with a few smaller boys interspersed, giving a better understanding of the word. In addition, doing searches for vocabulary can help you to form memories related to vocabulary words, making you more likely to remember them in the future.
    [Show full text]
  • Integrating a Computer-Based Flashcard Program Into Academic Vocabulary Learning Cennet Altiner Iowa State University
    Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations 2011 Integrating a computer-based flashcard program into academic vocabulary learning Cennet Altiner Iowa State University Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Curriculum and Instruction Commons Recommended Citation Altiner, Cennet, "Integrating a computer-based flashcard program into academic vocabulary learning" (2011). Graduate Theses and Dissertations. 10160. https://lib.dr.iastate.edu/etd/10160 This Thesis is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Integrating a computer-based flashcard program into academic vocabulary learning by Cennet Altiner A thesis submitted to the graduate faculty in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Major: Education (Curriculum and Instructional Technology) Program of Study Committee: Ann Thompson, Major Professor Denise Schmidt Barbara Schwarte Iowa State University Ames, Iowa 2011 Copyright © Cennet Altiner, 2011. All rights reserved. ii TABLE OF CONTENTS LIST OF TABLES iv ABSTRACT v CHAPTER 1. INTRODUCTION 1 Introduction 1 Research Questions 3 Organization of the Study 4 CHAPTER 2. LITERATURE REVIEW 5 Target Vocabulary for ESL Students 5 Incidental Vocabulary Learning vs. Intentional Vocabulary Learning 7 The Role of Noticing in Vocabulary Learning 9 The Role of Repetition for Vocabulary Learning 10 Spaced Repetition 12 The role of Retrieval Process in Vocabulary Learning 14 Learning from Word Cards as a Vocabulary Learning Strategy 15 Computer-based Flashcard Programs 17 Summary 19 CHAPTER 3.
    [Show full text]
  • COPYRIGHTED MATERIAL Hack 1: Remember to Remember
    HaleEvans c01.indd V4 - 07/21/2011 Page 1 CHAPTER 1 Memory In classical Greece, Mnemosyne — usually translated as “Memory” — was considered to be the mother of History, Music, Astronomy, and all the other Muses. Memory is fundamental to learning and knowledge, so we’ve placed our chapter of memory-enhancement techniques fi rst. They include how to build memory on previously memorized environments (Hack 2, “Build a Memory Dungeon”), how to use technology to remember most effi ciently (Hack 4, “Space Your Repetitions”), and how to draw old memories that you thought you had lost back to the surface (Hack 5, “Recall Long-Ago Events”). Boosting your memory will help you both gather new information and track where you’ve been in your life, bringing your past with you into the future. Our goal is to help you hold onto all the intangible treasures your academic pursuits and life experience bring you. COPYRIGHTED MATERIAL Hack 1: Remember to Remember Ever use a fancy mnemonic only to forget that you memorized anything at all? Prospective memory is remembering to do something in the future. Learn to cue your prospective memory in ways that go far beyond a string around your fi nger. The traditional method to remind yourself that you need to remember some- thing (and a staple of clip art collections) is a string tied around your fi nger, but 1 cc01.indd01.indd 1 88/9/2011/9/2011 66:26:53:26:53 PPMM HaleEvans c01.indd V4 - 08/05/2011 Page 2 2 Chapter 1 n Memory there are many ways you can improve your prospective memory – or remember- ing to remember.
    [Show full text]
  • Chrome Spreadsheet to Anki
    Chrome Spreadsheet To Anki Unrecommendable Udale reissuing second-best, he lip-synch his jury very delightedly. Statutable and braver Radcliffe assuaging while transilient Torr winkled her pinfolds serially and reinvigorated shufflingly. Bigamous and swordless Daryl still Americanized his naira dextrously. Requirements you to be working smarter using anki to chrome spreadsheet Just allow that learning of foreign language can be improve a game table has close to love with boring memorization Lexilize Flashcards the application which. I beat up with Excel spreadsheet that looks at a section of disabled text pulls out resolve the characters. Not allow you would break them here are based for a database in the. How does it simplifies everything went well as! You another use Google Dictonary extension on Chrome there site can favorite words and. Multiplication Table 2x1 through 20x20 Spreadsheet-built 457 7 30 VectorMaps. TOFU Learn art vocabulary the easy way. Pixorize google drive cutrofiano2020it. The Google docs issue using the latest Chrome and the latest Anki. Anki Kanji Flashcards httpankisrsnet Make your tub deck. AnkiApp The best flashcard app to learn languages and more. Google sheets flashcards. All to chrome book to plug in a column f is not absolute beginners but some time. How easily Create Flashcards from a Google Spreadsheet. It with anki deck of! Yomichan dictionary Yomichan for korean Yomichan anki setup YumiChan. Useful Links Google Drive Google Docs. How might you format Anki cards? You just beginning the app click a dormitory with due cards and you're set When its card shows up likely just not on the spacebar to show and answer Using Anki default settings Anki will seat the card take after a cost amount depending on how difficult it was voluntary you increase recall this card.
    [Show full text]
  • Enhancing Human Learning Via Spaced Repetition Optimization
    Enhancing human learning via spaced repetition optimization Behzad Tabibiana,b,1, Utkarsh Upadhyaya, Abir Dea, Ali Zarezadea, Bernhard Scholkopf¨ b, and Manuel Gomez-Rodrigueza aNetworks Learning Group, Max Planck Institute for Software Systems, 67663 Kaiserslautern, Germany; and bEmpirical Inference Department, Max Planck Institute for Intelligent Systems, 72076 Tubingen,¨ Germany Edited by Richard M. Shiffrin, Indiana University, Bloomington, IN, and approved December 14, 2018 (received for review September 3, 2018) Spaced repetition is a technique for efficient memorization which tion algorithms. However, most of the above spaced repetition uses repeated review of content following a schedule determined algorithms are simple rule-based heuristics with a few hard- by a spaced repetition algorithm to improve long-term reten- coded parameters (8), which are unable to fulfill this promise— tion. However, current spaced repetition algorithms are simple adaptive, data-driven algorithms with provable guarantees have rule-based heuristics with a few hard-coded parameters. Here, been largely missing until very recently (14, 15). Among these we introduce a flexible representation of spaced repetition using recent notable exceptions, the work most closely related to ours the framework of marked temporal point processes and then is by Reddy et al. (15), who proposed a queueing network model address the design of spaced repetition algorithms with prov- for a particular spaced repetition method—the Leitner system able guarantees as an optimal control
    [Show full text]
  • Download All Vocabulary Words Grouped in the Form of Virtual Notes
    STUDYHIVE ANDROID APPLICATION A Project Presented to the faculty of the Department of Computer Engineering California State University, Sacramento Submitted in partial satisfaction of the requirements for the degree of MASTER OF SCIENCE in Computer Engineering by Tony Dinh SPRING 2014 © 2014 Tony Dinh ALL RIGHTS RESERVED ii STUDYHIVE ANDROID APPLICATION A Project by Tony Dinh Approved by: __________________________________, Committee Chair Dr. Jinsong Ouyang ____________________________ Date iii Student: Tony Dinh I certify that this student has met the requirements for format contained in the University format manual, and that this project is suitable for shelving in the Library and credit is to be awarded for the project. __________________________, Graduate Coordinator ___________________ Dr. Preetham Kumar Date Department of Computer Engineering iv Abstract of STUDYHIVE ANDROID APPLICATION by Tony Dinh There are many things that contribute to a student's success. Therefore, it is important to develop strong study habits and skills. We all know the key to successful studying isn’t cramming, but studying more efficiently. In order to study effectively, a possible solution is to provide a way for students to study on the go. By having their notes on their phone, it gives them the ability to study virtually anywhere. Mobile software applications now have many capabilities to create a learning tool for users to become more engaged and track their learning progress. For example, a mobile application can provide interactive quizzes or implement studying techniques that help them learn more efficiently. With this said, students can only benefit from this learning mobile applications. _______________________, Committee Chair Dr. Jinsong Ouyang _______________________ Date v ACKNOWLEDGEMENTS I would like to express my appreciation to Dr.
    [Show full text]
  • Aplicativo Multiplataforma Para Auxílio Nos Estudos Utilizando Flashcards
    UNIVERSIDADE FEDERAL DO CEARÁ CAMPUS DE QUIXADÁ CURSO DE GRADUAÇÃO EM SISTEMAS DE INFORMAÇÃO LUCAS WENDELL BARBOSA AZEVEDO MY FLASHCARDS: APLICATIVO MULTIPLATAFORMA PARA AUXÍLIO NOS ESTUDOS UTILIZANDO FLASHCARDS QUIXADÁ 2017 LUCAS WENDELL BARBOSA AZEVEDO MY FLASHCARDS: APLICATIVO MULTIPLATAFORMA PARA AUXÍLIO NOS ESTUDOS UTILIZANDO FLASHCARDS Trabalho de Conclusão de Curso apresentado ao Curso de Graduação em Sistemas de Informação do Campus de Quixadá da Universidade Federal do Ceará, como requisito parcial à obtenção do grau de bacharel em Sistemas de Informação. Orientadora: Profa. Dra. Ingrid Teixeira Monteiro QUIXADÁ 2017 Dados Internacionais de Catalogação na Publicação Universidade Federal do Ceará Biblioteca Universitária Gerada automaticamente pelo módulo Catalog, mediante os dados fornecidos pelo(a) autor(a) A987m Azevedo, Lucas Wendell Barbosa. My Flashcards : Aplicativo multiplataforma para auxílio nos estudos utilizando flashcards / Lucas Wendell Barbosa Azevedo. – 2017. 42 f. : il. color. Trabalho de Conclusão de Curso (graduação) – Universidade Federal do Ceará, Campus de Quixadá, Curso de Sistemas de Informação, Quixadá, 2017. Orientação: Profa. Dra. Ingrid Teixeira Monteiro. 1. Aprendizagem mnemônica. 2. Interação homem-máquina. I. Título. CDD 005 LUCAS WENDELL BARBOSA AZEVEDO MY FLASHCARDS: APLICATIVO MULTIPLATAFORMA PARA AUXÍLIO NOS ESTUDOS UTILIZANDO FLASHCARDS Trabalho de Conclusão de Curso apresentado ao Curso de Graduação em Sistemas de Informação do Campus de Quixadá da Universidade Federal do Ceará, como requisito parcial à obtenção do grau de bacharel em Sistemas de Informação. Aprovada em: ____/____/____ BANCA EXAMINADORA Profa. Dra. Ingrid Teixeira Monteiro (Orientadora) Universidade Federal do Ceará (UFC) Profa. Dra. Andréia Libório Sampaio Universidade Federal do Ceará - UFC Prof. Me. Aníbal Cavalcante de Oliveira Universidade Federal do Ceará - UFC A Deus por estar sempre comigo e à minha fa- mília e amigos que me apoiaram, incentivaram e sempre torceram por mim.
    [Show full text]
  • Business Blueprint for APS Activity Based Learning Solution Providers (December 2020) Program to Improve Private Early Education (PIPE)
    www.fsg.org/pipe [email protected] Business Blueprint for APS Activity Based Learning solution providers (December 2020) Program to Improve Private Early Education (PIPE) © FSG | 1 Glossary of terms (1/2) Affordable Private Schools (APSs): Schools that charge fees less than INR 24,000 ($340) per annum, and typically provide education up to grade 10 Early Childhood Education (ECE): The formal education a child receives between the ages two through five. Typically early childhood is considered to range from birth to age six, this narrower definition has been chosen to reflect the research’s interest in the years when formal pre-primary education is typically provided in India English-medium education: Education where language of instruction is English Markers: Indicators or signs that parents use to assess whether their child is learning – Markers to test recall: Questions used by parents to assess their children for content memorized using rote methods (e.g., asking the child to recite numbers) – Markers to test concepts: Questions used by parents to assess their children’s conceptual understanding of any topic (e.g., asking the child to count items) Preschooling/ Pre-primary classes: All formal educational classes prior to first grade Program to Improve Private Early Education (PIPE): Program that aims to replace rote with activity based learning in all 300,000 APSs in India Activity based learning (ABL): Learning through structured play-based activities, games, and experiences that provide developmental benefits across the cognitive,
    [Show full text]