The Credit Problem in parametric stress
A probabilistic approach
Bibliographic Data
| ID | 22130866 |
|---|---|
| Authors | Aleksei Ioulevitch Nazarov (0000-0002-6682-6687, Utrecht University), Gaja Jarosz (0000-0002-8606-4704, University of Massachusetts Amherst) |
| Year | 2021 |
| Volume | 6 |
| Issue | 1 |
| Publication date | 2021-11-18 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Glossa a journal of general linguistics (JOURNAL) |
| Journal identifiers | ISSN: 2397-1835 • E-ISSN: 2397-1835 |
| Publisher | Open Library of the Humanities (PUBLISHER) |
| DOI | 10.16995/glossa.5884 |
| OpenAlex | W3215964486 |
| Language | EN |
| Citations received | 3 |
| References cited | 31 |
In this paper, we introduce a novel domain-general, statistical learning model for P&P grammars: the Expectation Driven Parameter Learner (EDPL). We show that the EDPL provides a mathematically principled solution to the Credit Problem (Dresher 1999). We present the first systematic tests of the EDPL and an existing and closely related model, the Naïve Parameter Learner (NPL), on a full stress typology, the one generated by Dresher & Kaye’s (1990) stress parameter framework. This framework has figured prominently in the debate about the necessity of domain-specific mechanisms for learning of parametric stress. The essential difference between the two learning models is that the EDPL incorporates a mechanism that directly tackles the Credit Problem, while the NPL does not. We find that the NPL fails to cope with the ambiguity of this stress system both in terms of learning success and data complexity, while the EDPL performs well on both metrics. Based on these results, we argue that probabilistic inference provides a viable domain-general approach to parametric stress learning, but only when learning involves an inferential process that directly addresses the Credit Problem. We also present in-depth analyses of the learning outcomes, showing how learning outcomes depend crucially on the structural ambiguities posited by a particular phonological theory, and how these learning difficulties correspond to typological gaps
Algorithmic learning theory · Ambiguity · Inference · Linguistics · Machine learning · Natural language processing · Parametric statistics · Probabilistic logic · Rule-based machine translation · Statistics · Computer Science · Language and cultural evolution · Mathematics · Natural Language Processing Techniques · Topic Modeling · Artificial Intelligence
Learnability in Optimality Theory
A computational learning model for metrical phonology
Maximum Likelihood from Incomplete Data Via the EM Algorithm
Slavic Prosody
The Handbook of Phonological Theory
Lectures on Government and Binding
How abstract is more abstract? Learning abstract underlying representations
Learning biases in opaque interactions
Bruce Tesar and Paul Smolensky (2000). Learnability in Optimality Theory . Cambridge, Mass
Learning with hidden structure in Optimality Theory and Harmonic Grammar
Knowledge and Learning in Natural Language
A Grammatical Sketch of Ossetic
Parsing to Learn
Computational Modeling of Phonological Learning
Learnability in Optimality Theory
Computational Phonology - Part I
Computational Phonology - Part II
Structure and Substance in Artificial-phonology Learning, Part I
Structure and Substance in Artificial-Phonology Learning, Part II
Typological Universals as Reflections of Biased Learning
Empirical Tests of the Gradual Learning Algorithm
Charting the Learning Path
Learnability Shapes Typology
| Citations per year | 1 |
|---|---|
| Citation span | 2023 - 2025 (3) |
| Highly cited | No |