Filters








42,208 Hits in 4.7 sec

An empirical study of learning and forgetting constraints

Ian P. Gent, Ian Miguel, Neil C.A. Moore
2012 AI Communications  
We conduct a major empirical investigation into the overheads introduced by unbounded constraint learning in CSP. To the best of our knowledge, this is the first published study in either CSP or SAT.  ...  The first is that a small percentage of learnt constraints do most propagation. While this is conventional wisdom, it has not previously been the subject of empirical study.  ...  Finally, we performed an empirical survey of several simple techniques for forgetting constraints in g-learning, and found that they are extremely effec-tive in making the learning solver more robust and  ... 
doi:10.3233/aic-2012-0524 fatcat:pmnlmyuz7jesrml3cl2lpg7jgu

Page 579 of The Journal of the Operational Research Society Vol. 56, Issue 5 [page]

2005 The Journal of the Operational Research Society  
In studying the effects of task complexity on increases, individual forgetting rates, various researchers have con- gher © cluded that higher task complexities result in hi . : 3.9.13 forgetting rates,  ...  DA Nembhard and N Osothsilp—Learning and forgetting-based worker selection wherein higher task complexity is associated with lower mean initial expertise, p.  ... 

Learning-forgetting independence, unidimensional memory models, and feature models: Comment on Bogartz (1990)

Geoffrey R. Loftus, Donald Bamber
1990 Journal of Experimental Psychology. Learning, Memory and Cognition  
We demonstrate what constraints must be placed on this model to make learning and forgetting rate independent by Loftus's and by Bogartz's definitions.  ...  Second, to better understand the constraints on memory mechanisms dictated by the mathematics of the models under consideration, we develop a simple but general feature model of learning and forgetting  ...  means of studying the dependence of forgetting rates on amount of learning" (p. 143).  ... 
doi:10.1037/0278-7393.16.5.916 pmid:2147446 fatcat:3qti6lm36zdxlhlpswbc75bkfq

Is Class-Incremental Enough for Continual Learning? [article]

Andrea Cossu, Gabriele Graffieti, Lorenzo Pellegrini, Davide Maltoni, Davide Bacciu, Antonio Carta, Vincenzo Lomonaco
2021 arXiv   pre-print
Each scenario defines the constraints and the opportunities of the learning environment.  ...  The ability of a model to learn continually can be empirically assessed in different continual learning scenarios.  ...  Continual learning for recurrent neural networks: An empirical evaluation.  ... 
arXiv:2112.02925v1 fatcat:42g3yyq5rrabrljsyxodccz2le

Online Continual Learning under Extreme Memory Constraints [article]

Enrico Fini, Stéphane Lathuilière, Enver Sangineto, Moin Nabi, Elisa Ricci
2022 arXiv   pre-print
In this paper, we introduce the novel problem of Memory-Constrained Online Continual Learning (MC-OCL) which imposes strict constraints on the memory overhead that a possible algorithm can use to avoid  ...  catastrophic forgetting.  ...  This work was carried out under the "Vision and Learning joint Laboratory" between FBK and UNITN.  ... 
arXiv:2008.01510v3 fatcat:nm3fsb5cxrgbletbchidzp6q5e

Putting the learning curve in context

J. Bradley Morrison
2008 Journal of Business Research  
Second, the model includes a budget constraint on time that forces a choice between an old and a new way to achieve the output.  ...  In implementation of change, people learn new ways of doing things, develop new skills, and adopt new organizational routines.  ...  Empirical study is also needed of both the dynamics of forgetting and the character of learning curves at low levels of experience.  ... 
doi:10.1016/j.jbusres.2007.11.009 fatcat:3t7qimqkszey7pg2pjfexx6uri

Learning and forgetting-based worker selection for tasks of varying complexity

D A Nembhard, N Osothsilp
2005 Journal of the Operational Research Society  
The effects of task complexity and experience on learning and forgetting: a field study.  ...  The learning curve: Historical review and comprehensive survey. Decision Sci 10: 302-328. Nembhard DA and Osothsilp N (2001). An empirical compar- ison of forgetting models.  ... 
doi:10.1057/palgrave.jors.2601842 fatcat:cpcbzj2cdbccvfxjj3euwr373q

Large-scale randomized experiment reveals machine learning helps people learn and remember more effectively [article]

Utkarsh Upadhyay and Graham Lancashire and Christoph Moser and Manuel Gomez-Rodriguez
2020 arXiv   pre-print
After controlling for the length and frequency of study, we find that learners whose study sessions are optimized using machine learning remember the content over ∼67 alternative heuristics.  ...  In this work, rather than replacing humans, we focus on unveiling the potential of machine learning to improve how people learn and remember factual material.  ...  We thank Robert West, Klein Lars Henning, Roland Aydin, and Behzad Tabibian for helpful conversations.  ... 
arXiv:2010.04430v1 fatcat:fmhaogmhjzdl5k2jtorz3o3jaq

A flow shop batch scheduling and operator assignment model with time-changing effects of learning and forgetting to minimize total actual flow time

Dwi Kurniawan, Andi Cakravastia Raja, Suprayogi Suprayogi, Abdul Hakim Halim
2020 Journal of Industrial Engineering and Management  
times as operators experience different degree of learning and forgetting.  ...  alternative operators but have not considered learning and forgetting, or have considered learning and forgetting but only in a single-stage system and without considering alternative operators.  ...  effects of learning and forgetting studied in Jaber and Bonney (1996) and Yusriski et al. (2015) .  ... 
doi:10.3926/jiem.3153 fatcat:3jg33d5lijhfposj4niu56fswa

Enhancing human learning via spaced repetition optimization

Behzad Tabibian, Utkarsh Upadhyay, Abir De, Ali Zarezade, Bernhard Schölkopf, Manuel Gomez-Rodriguez
2019 Proceedings of the National Academy of Sciences of the United States of America  
Here, we introduce a flexible representation of spaced repetition using the framework of marked temporal point processes and then address the design of spaced repetition algorithms with provable guarantees  ...  We perform a large-scale natural experiment using data from Duolingo, a popular language-learning online platform, and show that learners who follow a reviewing schedule determined by our algorithm memorize  ...  capturing particular learning goals and hard constraints on the number of events.  ... 
doi:10.1073/pnas.1815156116 pmid:30670661 pmcid:PMC6410796 fatcat:noqtkjoxnzhk5ockrvmlck47ce

Interference Produces Different Forgetting Rates for Implicit and Explicit Knowledge

Ricardo Tamayo, Peter A. Frensch
2007 Experimental Psychology  
However, the findings add an important empirical constraint to the existing literature and make clear that the most simple forms of the singlesystem theory cannot be correct because they cannot explain  ...  That is, the empirically observed dissociation in the time pattern of RTs and d's has been observed in at least two independent studies thus far; the obtained pattern is not unique to our study.  ...  Appendix Grammatical and Ungrammatical Strings Used in Experiments 1 and 2.  ... 
doi:10.1027/1618-3169.54.4.304 pmid:17953151 fatcat:w7jz6wjkovg65pgvxkfajkepgq

Gradient Regularized Contrastive Learning for Continual Domain Adaptation [article]

Peng Su, Shixiang Tang, Peng Gao, Di Qiu, Ni Zhao, Xiaogang Wang
2020 arXiv   pre-print
To better understand this issue, we study the problem of continual domain adaptation, where the model is presented with a labeled source domain and a sequence of unlabeled target domains.  ...  There are two major obstacles in this problem: domain shifts and catastrophic forgetting. In this work, we propose Gradient Regularized Contrastive Learning to solve the above obstacles.  ...  To overcome "catastrophic forgetting", we construct an additional set of constraints in which each constraint is imposed to enforce the classification loss of each domain-specific memory never increasing  ... 
arXiv:2007.12942v1 fatcat:6co6cmangngxznhsno4nmvngwy

Efficient Lifelong Learning with A-GEM [article]

Arslan Chaudhry, Marc'Aurelio Ranzato, Marcus Rohrbach, Mohamed Elhoseiny
2019 arXiv   pre-print
Third, we propose an improved version of GEM (Lopez-Paz & Ranzato, 2017), dubbed Averaged GEM (A-GEM), which enjoys the same or even better performance as GEM, while being almost as computationally and  ...  In lifelong learning, the learner is presented with a sequence of tasks, incrementally building a data-driven prior which may be leveraged to speed up learning of a new task.  ...  Unfortunately, this inner loop optimization becomes prohibitive when the size of M and the number of tasks is large, see Tab. 7 in Appendix for an empirical analysis.  ... 
arXiv:1812.00420v2 fatcat:r3qqrgpcvjbu3ks3lrsmajunxe

Unbounded Human Learning

Siddharth Reddy, Igor Labutov, Siddhartha Banerjee, Thorsten Joachims
2016 Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '16  
In the study of human learning, there is broad evidence that our ability to retain information improves with repeated exposure and decays with delay since last exposure.  ...  Finally, we empirically evaluate our queueing model through a Mechanical Turk experiment, verifying a key qualitative prediction of our model: the existence of a sharp phase transition in learning outcomes  ...  ACKNOWLEDGEMENTS We thank Peter Bienstman and the Mnemosyne project for making their data set publicly available.  ... 
doi:10.1145/2939672.2939850 dblp:conf/kdd/ReddyLBJ16 fatcat:47kl7davrzhcddoe7lamcy4fyy

Gradient Regularized Contrastive Learning for Continual Domain Adaptation [article]

Shixiang Tang, Peng Su, Dapeng Chen, Wanli Ouyang
2021 arXiv   pre-print
The obstacles in this problem are both domain shift and catastrophic forgetting. We propose Gradient Regularized Contrastive Learning (GRCL) to solve the obstacles.  ...  To better understand this issue, we study the problem of continual domain adaptation, where the model is presented with a labelled source domain and a sequence of unlabelled target domains.  ...  Target Memorization Constraint An essential problem in continual domain adaptation is catastrophic forgetting.  ... 
arXiv:2103.12294v1 fatcat:kycqokas6rahph2u4adpql2fmy
« Previous Showing results 1 — 15 out of 42,208 results