Filters








639,294 Hits in 7.5 sec

Acquiring Recursive Concepts with Explanation-Based Learning

Jude W. Shavlik
1989 International Joint Conference on Artificial Intelligence  
Explanation-based generalization algorithms need to generalize the structure of their explanations.  ...  This is necessary in order to acquire concepts where a recursive or iterative process is implicitly represented in the explanation by a fixed number of applications.  ...  His approach can acquire recursive and disjunctive concepts, as well as learn from multiple examples.  ... 
dblp:conf/ijcai/Shavlik89 fatcat:p34z2ryuy5cq5jhk7hprqa7aru

Acquiring recursive and iterative concepts with explanation-based learning

Iude W. Shavlik
1990 Machine Learning  
The BAGGER2 algorithm learns recursive and iterative concepts, integrates results from multiple examples, and extracts useful subconcepts during generalization.  ...  Most research in explanation-based learning involves relaxing constraints on the variables in the explanation of a specific example, rather than generalizing the graphical structure of the explanation  ...  The SOAR system [Laird et al, 1986] would seem to acquire a number of concepts that together are slightly more general.  ... 
doi:10.1007/bf00115894 fatcat:qufqu5zbnfcj3kdy45n43k42z4

Abductive explanation-based learning: A solution to the multiple inconsistent explanation problem

William W. Cohen
1992 Machine Learning  
Domain theories which suffer from the multiple inconsistent explanation problem can occur in many different contexts, such as when some information is missing and must be assumed: since such assumptions  ...  This paper proposes an extension of explanation-based learning, called abductive explanation-based learning (A-EBL) which solves the multiple inconsistent explanation problem by using set covering techniques  ...  Standard explanation-based systems cannot handle the multiple explanation problem.  ... 
doi:10.1007/bf00992863 fatcat:5wquk237n5d4jmosh22ajojksq

Probabilistic Explanation Based Learning [chapter]

Angelika Kimmig, Luc De Raedt, Hannu Toivonen
2007 Lecture Notes in Computer Science  
Explanation based learning produces generalized explanations from examples.  ...  These probabilistic and generalized explanations can then be used to discover similar examples and to reason by analogy.  ...  This research has been partially supported by IQ (European Union Project IST-FET FP6-516169), Research Foundation-Flanders (FWO-Vlaanderen), Tekes and Humboldt foundation.  ... 
doi:10.1007/978-3-540-74958-5_19 fatcat:fopeklsc2nepbbpaapkfn6wkwq

CLUES: A Benchmark for Learning Classifiers using Natural Language Explanations [article]

Rakesh R Menon, Sayan Ghosh, Shashank Srivastava
2022 arXiv   pre-print
It contains crowdsourced explanations describing real-world tasks from multiple teachers and programmatically generated explanations for the synthetic tasks.  ...  To model the influence of explanations in classifying an example, we develop ExEnt, an entailment-based model that learns classifiers using explanations.  ...  Thus, the tasks in CLUES-Real contain explanations from multiple teachers and multiple students corresponding to a teacher.  ... 
arXiv:2204.07142v1 fatcat:cey5umslvvhgjjtpa2vuclx4aa

Explanation-based generalization: A unifying view

Tom M. Mitchell, Richard M. Keller, Smadar T. Kedar-Cabelli
1986 Machine Learning  
The EBG .method is illustrated in the context of several example problems, and used to contrast several existing systems for explanation-based generalization.  ...  The perspective on explanation-based generalization afforded by this general method is also used to identify open research problems in this area.  ...  Acknowledgments The perspective on Explanation-Based Generalization reported here has arisen from discussions over a period of time with a number of people in the Machine Learning Group at Rutgers and  ... 
doi:10.1007/bf00116250 fatcat:nngighwmczgkvhvpkkdxevyls4

The effects of self-explaining when learning with text or diagrams

S Ainsworth
2003 Cognitive Science  
To discover why diagrams can promote the self-explanation effect, results are interpreted with reference to the multiple differences in the semantic, cognitive and affective properties of the texts and  ...  Diagrams students also generated significantly more self-explanations that text students. Furthermore, the benefits of self-explaining were much greater in the diagrams condition.  ...  Number of self-explanations was significantly negatively correlated with time spent learning and number of words.  ... 
doi:10.1016/s0364-0213(03)00033-8 fatcat:xyagormagfhqhd2u4c2mhab6am

The effects of self-explaining when learning with text or diagrams

Shaaron Ainsworth, Andrea Th Loizou
2003 Cognitive Science  
To discover why diagrams can promote the self-explanation effect, results are interpreted with reference to the multiple differences in the semantic, cognitive and affective properties of the texts and  ...  Diagrams students also generated significantly more self-explanations that text students. Furthermore, the benefits of self-explaining were much greater in the diagrams condition.  ...  Number of self-explanations was significantly negatively correlated with time spent learning and number of words.  ... 
doi:10.1207/s15516709cog2704_5 fatcat:ea3fyhfowvfstdakbos5dsn6mu

An Explanation-based Approach to Generalizing Number

Jude W. Shavlik, Gerald DeJong
1987 International Joint Conference on Artificial Intelligence  
An approach to generalizing number in explanation-based learning is presented.  ...  This issue has been largely ignored in previous explanation-based learning research.  ...  This paper presents an approach to generalizing to N in explanation-based learning.  ... 
dblp:conf/ijcai/ShavlikD87 fatcat:vjow47pfxng3dfd7uklzpjvzha

Induction over the unexplained: Using overly-general domain theories to aid concept learning

Raymond J. Mooney
1993 Machine Learning  
This paper describes and evaluates an approach to combining empirical and explanation-based learning called Induction Over the Unexplained (IOU).  ...  Empirical results show that IOU is effective at refining overly-general domain theories and that it learns more accurate concepts from fewer examples than a purely empirical approach.  ...  This work was supported by the NASA Ames Research Center through grant number NCC 2-629. Notes 1.  ... 
doi:10.1007/bf00993482 fatcat:gqruhmzonjhhtfxjv33kjt2jky

Learning search control knowledge: An explanation-based approach

Roland J. Zito-Wolf
1991 Machine Learning  
An algorithm, Explanation-Based Sepcialization (EBS), is presented and proved correct. • The application of EBL is extended to meta-level concepts, concepts acquired across multiple examples, and learning  ...  PRODIGY uses Explanation-Based Learning (EBL) to distill effective search-control knowledge from experience.  ...  Acknowledgments Many thanks to Glenn Iba for discussions on macro-learning, to Don Smith and Tim Hickey for extensive help with partial evaluation, and to Rick Alterman for suggesting this project and  ... 
doi:10.1007/bf00114164 fatcat:hki6sfjiknfsnltvrbeaiu6t3e

Automated Personalized Feedback Improves Learning Gains in an Intelligent Tutoring System [article]

Ekaterina Kochmar, Dung Do Vu, Robert Belfer, Varun Gupta, Iulian Vlad Serban, Joelle Pineau
2020 arXiv   pre-print
We utilize state-of-the-art machine learning and natural language processing techniques to provide the students with personalized hints, Wikipedia-based explanations, and mathematical hints.  ...  learning outcomes and in the subjective evaluation of the feedback.  ...  A Decision Tree classifier is trained on a dataset of positive and negative examples to evaluate the quality of a Wikipedia-based explanation using a number of linguistically-motivated features.  ... 
arXiv:2005.02431v2 fatcat:z7v7uwlg35dh7e4xz7pynjrjla

Integrating Inductive Neural Network Learning and Explanation-Based Learning

Sebastian Thrun, Tom M. Mitchell
1993 International Joint Conference on Artificial Intelligence  
Explanations are constructed by chaining together inferences from multiple neural networks.  ...  We present here a learning method that combines explanation-based learning from a previously learned approximate domain theory, together with inductive learning from observations.  ...  Acknowledgment We thank Ryusuke Masuoka for his invaluable help in refining the EBNN algorithm and code.  ... 
dblp:conf/ijcai/ThrunM93 fatcat:uqe4v4hr5vapjaoqv6hn5dwmeq

Automated Personalized Feedback Improves Learning Gains in An Intelligent Tutoring System [chapter]

Ekaterina Kochmar, Dung Do Vu, Robert Belfer, Varun Gupta, Iulian Vlad Serban, Joelle Pineau
2020 Lecture Notes in Computer Science  
We utilize state-of-the-art machine learning and natural language processing techniques to provide the students with personalized hints, Wikipedia-based explanations, and mathematical hints.  ...  improvement in student learning outcomes and in the subjective evaluation of the feedback.  ...  A Decision Tree classifier is trained on a dataset of positive and negative examples to evaluate the quality of a Wikipedia-based explanation using a number of linguistically-motivated features.  ... 
doi:10.1007/978-3-030-52240-7_26 fatcat:ffys6k3pgfbs3nh2yx73lrg7yi

News and notes

1987 Machine Learning  
Support for the review was provided by the National Science Foundation under grant number IST-8510999 and by the Army Research Office under contract ARO DAAG29-84-K-0060.  ...  Kodratoff's future research concerns handling multiple explanations derived from a single example.  ...  Reed (University of Birmingham), Some Predictive Difficulties in Automatic Induction. characterized by the learning of general rules or hypotheses from multiple examples.  ... 
doi:10.1007/bf00058755 fatcat:c5ixjwwgb5fzrifunchwl2wbf4
« Previous Showing results 1 — 15 out of 639,294 results