A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
A Clustering and Demotion Based Algorithm for Inductive Learning of Default Theories
[article]
2021
arXiv
pre-print
Preserved Fulltext
We present a clustering- and demotion-based algorithm called Kmeans-FOLD to induce nonmonotonic logic programs from positive and negative examples. ...
Our experiments on the UCI dataset show that a combination of K-Means clustering and our demotion strategy produces significant improvement for datasets with more than one cluster of positive examples. ...
(v) extending our algorithm for multi-relational input datasets. Note that our work reported here that makes use of clustering and demotion to improve the FOLD algorithm is just the beginning. ...
arXiv:2109.12624v1
fatcat:6gkpo2yco5cgniiwr5wo7naowm
A Maximum Entropy Model of Phonotactics and Phonotactic Learning
2008
Linguistic Inquiry
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The study of phonotactics is a central topic in phonology. We propose a theory of phonotactic grammars and a learning algorithm that constructs such grammars from positive evidence. ...
We apply the model in a variety of learning simulations, showing that the learned grammars capture the distributional generalizations of these languages and accurately predict the findings of a phonotactic ...
For OT, the Constraint Demotion algorithm family of Tesar and Smolensky (1998, 2000) satisfies the first of these criteria in a sense, as it provably converges for the case of consistent input-output ...
doi:10.1162/ling.2008.39.3.379
fatcat:ft5axndkgvabdmnnkutsaqbiom
Optimality Theory and phonological acquisition
2004
Annual Review of Language Acquisition
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
We thank the editors of ARLA for helpful comments on an earlier version of this paper. ...
Acknowledgements We are grateful to the people who responded to our survey for their time and effort: Jessica Barlow, Katherine Demuth, Daniel Dinnsen, Dicky Gilbers, Steven Gillis, Bruce Hayes, Lise Menn ...
It is not bad in general for a learning algorithm to fail on certain input data. ...
doi:10.1075/arla.3.03boe
fatcat:ejblyl6birachdsykjek66nmpy
Participant association and emergent curriculum in a MOOC: can the community be the curriculum?
2016
Research in Learning Technology
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
We investigated how participants associated with each other and developed community in a Massive Open Online Course (MOOC) about Rhizomatic Learning (Rhizo14). ...
Our combination of thematic analysis of qualitative survey data with analysis of participant observation, activity data, archives and visualisation of SNS data enabled us to reach a deeper understanding ...
Acknowledgements We would like to thank Rhizo14 participants especially those who gave their time to complete the survey and answer our questions. ...
doi:10.3402/rlt.v24.29927
fatcat:zlqq54djlnby5digfdxq5bwfna
An empirical study of on-line models for relational data streams
2016
Machine Learning
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
In this paper, we investigate the adoption of a stream-based on-line learning approach to relational data. ...
To date, Inductive Logic Programming (ILP) systems have largely assumed that all data needed for learning have been provided at the onset of model construction. ...
Some part of this work was supported by a Ramanujan Fellowship from the Department of Science and Technology, Government of India. ...
doi:10.1007/s10994-016-5596-2
fatcat:hxumvsflobg4po4e2fzv6xsm3q
Loanword accentuation in Yanbian Korean: a weighted-constraints analysis
2013
Natural language & linguistic theory
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
These markedness constraints are weighted by the learning algorithm so that the weight hierarchy can achieve a more or less "faithful adaptation" of the source language. ...
Through a learning process, the original faithfulness constraints to the source language are demoted below relevant markedness constraints. ...
I also thank Adam Albright for sharing with me the program for the learning algorithm used in section 4.1, and Shigeto Kawahara for advice about statistics. ...
doi:10.1007/s11049-013-9211-y
fatcat:4lvfom2xvjdgrjzv6wb5mlfshu
Machine Learning in Automated Text Categorization
[article]
2001
arXiv
pre-print
Preserved Fulltext
The Internet Archive has a preservation copy of this work in our general collections.
The file type is application/pdf.
In the research community the dominant approach to this problem is based on machine learning techniques: a general inductive process automatically builds a classifier by learning, from a set of preclassified ...
of expert manpower, and straightforward portability to different domains. ...
Acknowledgements This paper owes a lot to the suggestions and constructive criticism of Norbert Fuhr and David Lewis. ...
arXiv:cs/0110053v1
fatcat:fmqocqebfnahlhbnpizcvo6kri
Bias in data‐driven artificial intelligence systems—An introductory survey
2020
Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In this survey, we focus on data-driven AI, as a large part of AI is powered nowadays by (big) data and powerful machine learning algorithms. ...
Therefore, it is necessary to move beyond traditional AI algorithms optimized for predictive performance and embed ethical and legal principles in their design, training, and deployment to ensure social ...
CONFLICT OF INTEREST The authors have declared no conflicts of interest for this article. ...
doi:10.1002/widm.1356
fatcat:hbdgabycvndpjmgn7rjzkkk5ma
Stochastic phonological knowledge: the case of Hungarian vowel harmony
2006
Phonology
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2006; you can also visit the original URL.
The file type is application/pdf.
We model the speakers' knowledge and intuitions with a grammar constructed under the dual listing/generation model of Zuraw (2000) , then show how the constraint rankings of this grammar can be learned ...
by algorithm. * * We would like to thank for helpful advice. ...
Both of our algorithms, Low Faithfulness Constraint Demotion and Biased Constraint Demotion, require a set of losing candidates for learning (the rankings are learned by comparing these losers with winning ...
doi:10.1017/s0952675706000765
fatcat:qehnihi5dffghjqjmfqqxwkwee
The Actuation Problem in Optimality Theory
[chapter]
2003
Studies in Natural Language and Linguistic Theory
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
We show that, whereas OT predicts this state of affairs, rule-based theories cannot account for the facts without imposing contradictory demands on acquisition theory. ...
It is therefore suggested that, as a theory of grammar, OT will play a secondary role in accounts of phonologization. ...
These criteria are compatible with noninnatist approaches such as Hayes's (1999) theory of inductive grounding, where the learner is equipped with an algorithm for constraint discovery that guarantees ...
doi:10.1007/978-94-010-0195-3_4
fatcat:fqxrhr7otzerffou3e4a2woalu
Bias in Data-driven AI Systems – An Introductory Survey
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In this survey, we focus on data-driven AI, as a large part of AI is powered nowadays by (big) data and powerful Machine Learning (ML) algorithms. ...
Therefore, it is necessary to move beyond traditional AI algorithms optimized for predictive performance and embed ethical and legal principles in their design, training and deployment to ensure social ...
Acknowledgement This work is supported by the project "NoBias -Artificial Intelligence without Bias", which has received funding from the European Union's Horizon 2020 research and innovation programme ...
arXiv:2001.09762v1
fatcat:is3pn4c2srgdfbjnpvbwodoura
Functional Representation of Prototypes in LVQ and Relevance Learning
[chapter]
2016
Advances in Intelligent Systems and Computing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The overall acceptance rate was 88% (63% for regular papers, 100% for compressed contributions and demonstration abstracts, and 91% for thesis abstracts). ...
(University of Kent), Frank van Harmelen (Vrije Universiteit Amsterdam), Hado van Hasselt (Google DeepMind), and Manuela Veloso (Carnegie Mellon University), a Research meets Business session, a panel ...
This work successfully builds on previous theories from the field of new media and linguistics. ...
doi:10.1007/978-3-319-28518-4_28
fatcat:uwxvq6txmrba3ajulmblafgh2a
Cognitive resource depletion in religious interactions
2013
Religion, Brain & Behavior
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
We explore the cognitive effects of three common features of religious interactions: (1) demand for the expressive suppression of emotion; (2) exposure to goaldemoted and causally opaque actions; and ( ...
Using a cognitive resource model of executive function, we argue that these three features affect the executive system in ways that limit the capacity for individual processing of religious events. ...
Many of our actions and choices are the result of subconscious algorithms and associational processing. ...
doi:10.1080/2153599x.2012.736714
fatcat:mvb7nhxsizc4jopt55dp67jcmu
On Evaluation Metrics in Optimality Theory
2016
Linguistic Inquiry
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
We develop an evaluation metric for Optimality Theory that allows a learner to induce a lexicon and a phonological grammar from unanalyzed surface forms. ...
final l of table 'table' is optional and can be deleted while that of parle 'speak' cannot. ...
Our goal has been to argue for MDL as a learning theory for OT, and Heinz and Idsardi's generalization is currently of no help in choosing among such theories. ...
doi:10.1162/ling_a_00210
fatcat:7cq2lygd6zhlfpln5rvikftbqm
Deletion or epenthesis? On the falsifiability of phonological universals
2015
Lingua
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2015; you can also visit the original URL.
The file type is application/pdf.
This paper presents a revised typology of consonant epenthesis and explores the theoretical implications of such a typology. ...
Through careful re-analysis, the basis for a proposed universal of coronal preference and dorsal avoidance is shown to be lacking in evidential support. ...
property of a learning algorithm). ...
doi:10.1016/j.lingua.2014.11.002
fatcat:xw6dsnecyvb2bbqs6bgnge6oxm
« Previous
Showing results 1 — 15 out of 143 results