Jeffrey Heinz
Dados Biográficos
| ID | 822553 |
|---|---|
| NOME | Jeffrey Heinz |
| PRENOMES | Jeffrey |
| SOBRENOME | Heinz |
| ASSINATURA | HEINZ J |
| AFILIAÇÕES | University of Delaware |
| ORCID | 0000-0002-5954-3195 |
| VERIFICADO | Sim |
| TOTAL DE OBRAS | 20 |
| TOTAL DE CITAÇÕES | 25 |
| TOTAL COMO AUTOR | 20 |
| TOTAL COMO EDITOR | 0 |
| PRIMEIRO ANO DE PUBLICAÇÃO | 2009 |
| ANO MAIS RECENTE DE PUBLICAÇÃO | 2026 |
| ÍNDICE H | 3 |
Interpreting state, with focus on metaphonic processes
Metaphony provides a unique window into the nature of the capacities of phonological grammars. The representational issues metaphonic processes highlight are exactly those core to computational analysis. Prominent examples of such processes, under various systems of phonological and morphological representation, are examined using algebraic methods, which describe the structure that arises when certain sequences exhibit the same behaviours. This …
Phonotactic Learning with Structure, Not Statistics
We provide empirical evidence against Wilson & Gallagher’s 2018 claim that statistics is necessary for phonotactic learning. We implement BUFIA, a feature-based nonstatistical learner and, using the same data and case study as Wilson & Gallagher, show that this non-statistical learner is equally successful at learning phonotactics as the Maximum Entropy-based learner they use. This counters their conclusion that non-statistical phonotactic learni…
The blueprint model of production
This article introduces the blueprint model of production (BMP), which characterises the phonetics–phonology interface in terms of typed functions. The standard modular feed-forward view to the interface is that the phonetic form of a lexical item is the output of a phonetic module which takes the output of a phonological module as its input. The central idea of the BMP is that the phonetic form is instead the output of a higher-order phonetics f…
Regular and polyregular theories of reduplication
We explore the generative capacity of morphological theories of reduplication. We computationally classify theories of reduplication using a hierarchy of string-to-string function classes. Reduplication as a process requires only the regular class of functions. We show that various morphological theories necessarily treat it as a more expressive polyregular function, while others maintain regularity. We discuss the significance of this formal res…
Incomplete neutralization and the blueprint model of production
The division of labor between the phonetic and phonological modules in generative linguistics has often been allocated such that phonology handles the discrete and symbolic aspects of a language's sound system, while the phonetics transforms the symbols either into, or from, continuously varying acoustic and/or articulatory representations. This type of architecture struggles with accounting for incomplete neutralization where two segments are ne…
Categorical Account of Gradient Acceptability of Word-Initial Polish Onsets
We examine how well categorical and probabilistic phonotactic learning models extract grammars which predict Polish speakers' acceptability judgments of words with varied initial consonant clusters. Polish is an especially interesting language to look at because of its rich inventory of sonority-sequencing defying consonant clusters, often as a result of yer-deletion. In line with results by Gorman (2013) and Durvasula (2020), we find that the ca…
No free lunch in linguistics or machine learning
Pater's (2019) target article proposes that neural networks will provide theories of learning that generative grammar lacks. We argue that his enthusiasm is premature since the biases of neural networks are largely unknown, and he disregards decades of work on machine learning and learnability. Learning biases form a two-way street: all learners have biases, and those biases constrain the space of learnable grammars in mathematically measurable w…
No free lunch in linguistics or machine learning
Pater’s (2019) target article proposes that neural networks will provide theories of learning that generative grammar lacks. We argue that his enthusiasm is premature since the biases of neural networks are largely unknown, and he disregards decades of work on machine learning and learnability. Learning biases form a two-way street: all learners have biases, and those biases constrain the space of learnable grammars in mathematically measurable w…
Input Strictly Local opaque maps
This paper gives a computational characterisation of opaque interactions in phonology. Specifically, a range of opaque interactions are shown to be Input Strictly Local (ISL) maps (Chandlee 2014), which are string-to-string functions that determine output based only on contiguous sequences of input symbols. Examples from Baković’s (2007) extended typology of counterfeeding, counterbleeding, self-destructive feeding, non-gratuitous feeding and cro…
Strict Locality and Phonological Maps
In this article, we identify Strict Locality as a strong computational property of a certain class of phonological maps from underlying to surface forms. We show that these maps can be modeled with Input Strictly Local functions, a previously undefined class of subregular relations. These functions extend the conception of locality from the Strictly Local formal languages (recognizers/acceptors) ( McNaughton and Papert 1971 , Rogers and Pullum 20…
Computational phonology today
This thematic issue almost did not happen. One of us (JH) was almost killed two days after the deadline for article submissions. As a pedestrian on a sidewalk minding his own business, he was struck by a car that ran a red light and lost control after a collision. So when we write that we are delighted to be writing this introduction, over one year later, we both really mean it
A formal analysis of Correspondence Theory
This paper provides a computational analysis of the complexity of GEN and Correspondence Theory in terms of the nature of the logic involved in their formulation. The first result of this analysis shows that the GEN function is not definable in Monadic Second Order (MSO) logic. Second, we show that the set of input-output Correspondence-theoretic candidates from a given underlying representation is definable in First Order (FO) logic, which is le…
Learning Repairs for Marked Structures
[Abstract not available]
Neutral vowels in sound-symbolic vowel harmony in Korean
The set of neutral vowels in the vowel-harmony pattern found in Modern Korean sound-symbolic words is traditionally considered to consist of two members: [i] and [ɨ]. Some authors have tentatively put forward the hypothesis that [u] and [y] are also neutral. A corpus study on the phonotactics of Korean sound-symbolic vowel harmony provides quantitative support both for the traditional understanding of [i] and [ɨ] as neutral vowels and for the ten…
Computational Phonology - Part I
Computational phonology approaches the study of sound patterns in the world's languages from a computational perspective. This article explains this perspective and its relevance to phonology. A restrictive, universal property of phonological patterns - they are regular - is established, and the hypothesis that they are subregular is presented. This article is intended primarily for phonologists who are curious about computational phonology, but …
Computational Phonology - Part II
Computational phonology studies sound patterns in the world's languages from a computational perspective. This article shows that the similarities between different generative theories outweigh the differences, and discusses stochastic grammars and learning models within phonology from a computational perspective. Also, it shows how the hypothesis that all sound patterns are subregular can be investigated, pointing the direction for future resear…
Modeling the contribution of phonotactic cues to the problem of word segmentation
How do infants find the words in the speech stream? Computational models help us understand this feat by revealing the advantages and disadvantages of different strategies that infants might use. Here, we outline a computational model of word segmentation that aims both to incorporate cues proposed by language acquisition researchers and to establish the contributions different cues can make to word segmentation. We present experimental results f…
Learning Long-Distance Phonotactics
This article shows that specific properties of long-distance phonotactic patterns derived from consonantal harmony patterns (Hansson 2001, Rose and Walker 2004) follow from a learner that generalizes only on the basis of the order of sounds, not the distance between them. The proposed learner is simple, efficient, and provably correct, and does not require an a priori notion of tier or projection (contra the model in Hayes and Wilson 2008); nor d…
On the role of locality in learning stress patterns
This paper presents a previously unnoticed universal property of stress patterns in the world's languages: they are, for small neighbourhoods, neighbourhood-distinct. Neighbourhood-distinctness is a locality condition defined in automata-theoretic terms. This universal is established by examining stress patterns contained in two typological studies. Strikingly, many logically possible – but unattested – patterns do not have this property. Not onl…
Evaluating the Complexity of Optimality Theory
Idsardi (2006) claims that Optimality Theory (OT; Prince and Smolensky 1993, 2004) is "in general computationally intractable" on the basis of a proof adapted from Eisner 1997a. We take issue with this conclusion on two grounds. First, the intractability result holds only in cases where the constraint set is not fixed in advance (contra usual definitions of OT), and second, the result crucially depends on a particular representation of OT grammar…
Learning Long-Distance Phonotactics
This article shows that specific properties of long-distance phonotactic patterns derived from consonantal harmony patterns (Hansson 2001, Rose and Walker 2004) follow from a learner that generalizes only on the basis of the order of sounds, not the distance between them. The proposed learner is simple, efficient, and provably correct, and does not require an a priori notion of tier or projection (contra the model in Hayes and Wilson 2008); nor d…
Strict Locality and Phonological Maps
In this article, we identify Strict Locality as a strong computational property of a certain class of phonological maps from underlying to surface forms. We show that these maps can be modeled with Input Strictly Local functions, a previously undefined class of subregular relations. These functions extend the conception of locality from the Strictly Local formal languages (recognizers/acceptors) ( McNaughton and Papert 1971 , Rogers and Pullum 20…
Computational Phonology - Part I
Computational phonology approaches the study of sound patterns in the world's languages from a computational perspective. This article explains this perspective and its relevance to phonology. A restrictive, universal property of phonological patterns - they are regular - is established, and the hypothesis that they are subregular is presented. This article is intended primarily for phonologists who are curious about computational phonology, but …
Evaluating the Complexity of Optimality Theory
Idsardi (2006) claims that Optimality Theory (OT; Prince and Smolensky 1993, 2004) is "in general computationally intractable" on the basis of a proof adapted from Eisner 1997a. We take issue with this conclusion on two grounds. First, the intractability result holds only in cases where the constraint set is not fixed in advance (contra usual definitions of OT), and second, the result crucially depends on a particular representation of OT grammar…
Computational Phonology - Part II
Computational phonology studies sound patterns in the world's languages from a computational perspective. This article shows that the similarities between different generative theories outweigh the differences, and discusses stochastic grammars and learning models within phonology from a computational perspective. Also, it shows how the hypothesis that all sound patterns are subregular can be investigated, pointing the direction for future resear…
On the role of locality in learning stress patterns
This paper presents a previously unnoticed universal property of stress patterns in the world's languages: they are, for small neighbourhoods, neighbourhood-distinct. Neighbourhood-distinctness is a locality condition defined in automata-theoretic terms. This universal is established by examining stress patterns contained in two typological studies. Strikingly, many logically possible – but unattested – patterns do not have this property. Not onl…
Evaluating the Complexity of Optimality Theory
Idsardi (2006) claims that Optimality Theory (OT; Prince and Smolensky 1993, 2004) is "in general computationally intractable" on the basis of a proof adapted from Eisner 1997a. We take issue with this conclusion on two grounds. First, the intractability result holds only in cases where the constraint set is not fixed in advance (contra usual definitions of OT), and second, the result crucially depends on a particular representation of OT grammar…
Modeling the contribution of phonotactic cues to the problem of word segmentation
How do infants find the words in the speech stream? Computational models help us understand this feat by revealing the advantages and disadvantages of different strategies that infants might use. Here, we outline a computational model of word segmentation that aims both to incorporate cues proposed by language acquisition researchers and to establish the contributions different cues can make to word segmentation. We present experimental results f…
Learning Long-Distance Phonotactics
This article shows that specific properties of long-distance phonotactic patterns derived from consonantal harmony patterns (Hansson 2001, Rose and Walker 2004) follow from a learner that generalizes only on the basis of the order of sounds, not the distance between them. The proposed learner is simple, efficient, and provably correct, and does not require an a priori notion of tier or projection (contra the model in Hayes and Wilson 2008); nor d…
Computational Phonology - Part I
Computational phonology approaches the study of sound patterns in the world's languages from a computational perspective. This article explains this perspective and its relevance to phonology. A restrictive, universal property of phonological patterns - they are regular - is established, and the hypothesis that they are subregular is presented. This article is intended primarily for phonologists who are curious about computational phonology, but …
Computational Phonology - Part II
Computational phonology studies sound patterns in the world's languages from a computational perspective. This article shows that the similarities between different generative theories outweigh the differences, and discusses stochastic grammars and learning models within phonology from a computational perspective. Also, it shows how the hypothesis that all sound patterns are subregular can be investigated, pointing the direction for future resear…
Neutral vowels in sound-symbolic vowel harmony in Korean
The set of neutral vowels in the vowel-harmony pattern found in Modern Korean sound-symbolic words is traditionally considered to consist of two members: [i] and [ɨ]. Some authors have tentatively put forward the hypothesis that [u] and [y] are also neutral. A corpus study on the phonotactics of Korean sound-symbolic vowel harmony provides quantitative support both for the traditional understanding of [i] and [ɨ] as neutral vowels and for the ten…
Learning Repairs for Marked Structures
[Abstract not available]
Computational phonology today
This thematic issue almost did not happen. One of us (JH) was almost killed two days after the deadline for article submissions. As a pedestrian on a sidewalk minding his own business, he was struck by a car that ran a red light and lost control after a collision. So when we write that we are delighted to be writing this introduction, over one year later, we both really mean it
A formal analysis of Correspondence Theory
This paper provides a computational analysis of the complexity of GEN and Correspondence Theory in terms of the nature of the logic involved in their formulation. The first result of this analysis shows that the GEN function is not definable in Monadic Second Order (MSO) logic. Second, we show that the set of input-output Correspondence-theoretic candidates from a given underlying representation is definable in First Order (FO) logic, which is le…
Input Strictly Local opaque maps
This paper gives a computational characterisation of opaque interactions in phonology. Specifically, a range of opaque interactions are shown to be Input Strictly Local (ISL) maps (Chandlee 2014), which are string-to-string functions that determine output based only on contiguous sequences of input symbols. Examples from Baković’s (2007) extended typology of counterfeeding, counterbleeding, self-destructive feeding, non-gratuitous feeding and cro…
Strict Locality and Phonological Maps
In this article, we identify Strict Locality as a strong computational property of a certain class of phonological maps from underlying to surface forms. We show that these maps can be modeled with Input Strictly Local functions, a previously undefined class of subregular relations. These functions extend the conception of locality from the Strictly Local formal languages (recognizers/acceptors) ( McNaughton and Papert 1971 , Rogers and Pullum 20…
No free lunch in linguistics or machine learning
Pater's (2019) target article proposes that neural networks will provide theories of learning that generative grammar lacks. We argue that his enthusiasm is premature since the biases of neural networks are largely unknown, and he disregards decades of work on machine learning and learnability. Learning biases form a two-way street: all learners have biases, and those biases constrain the space of learnable grammars in mathematically measurable w…
No free lunch in linguistics or machine learning
Pater’s (2019) target article proposes that neural networks will provide theories of learning that generative grammar lacks. We argue that his enthusiasm is premature since the biases of neural networks are largely unknown, and he disregards decades of work on machine learning and learnability. Learning biases form a two-way street: all learners have biases, and those biases constrain the space of learnable grammars in mathematically measurable w…
Incomplete neutralization and the blueprint model of production
The division of labor between the phonetic and phonological modules in generative linguistics has often been allocated such that phonology handles the discrete and symbolic aspects of a language's sound system, while the phonetics transforms the symbols either into, or from, continuously varying acoustic and/or articulatory representations. This type of architecture struggles with accounting for incomplete neutralization where two segments are ne…
Categorical Account of Gradient Acceptability of Word-Initial Polish Onsets
We examine how well categorical and probabilistic phonotactic learning models extract grammars which predict Polish speakers' acceptability judgments of words with varied initial consonant clusters. Polish is an especially interesting language to look at because of its rich inventory of sonority-sequencing defying consonant clusters, often as a result of yer-deletion. In line with results by Gorman (2013) and Durvasula (2020), we find that the ca…
Regular and polyregular theories of reduplication
We explore the generative capacity of morphological theories of reduplication. We computationally classify theories of reduplication using a hierarchy of string-to-string function classes. Reduplication as a process requires only the regular class of functions. We show that various morphological theories necessarily treat it as a more expressive polyregular function, while others maintain regularity. We discuss the significance of this formal res…
The blueprint model of production
This article introduces the blueprint model of production (BMP), which characterises the phonetics–phonology interface in terms of typed functions. The standard modular feed-forward view to the interface is that the phonetic form of a lexical item is the output of a phonetic module which takes the output of a phonological module as its input. The central idea of the BMP is that the phonetic form is instead the output of a higher-order phonetics f…
Interpreting state, with focus on metaphonic processes
Metaphony provides a unique window into the nature of the capacities of phonological grammars. The representational issues metaphonic processes highlight are exactly those core to computational analysis. Prominent examples of such processes, under various systems of phonological and morphological representation, are examined using algebraic methods, which describe the structure that arises when certain sequences exhibit the same behaviours. This …
Phonotactic Learning with Structure, Not Statistics
We provide empirical evidence against Wilson & Gallagher’s 2018 claim that statistics is necessary for phonotactic learning. We implement BUFIA, a feature-based nonstatistical learner and, using the same data and case study as Wilson & Gallagher, show that this non-statistical learner is equally successful at learning phonotactics as the Maximum Entropy-based learner they use. This counters their conclusion that non-statistical phonotactic learni…
Computer Science (17 obras) · Linguistics (14 obras) · Natural Language Processing Techniques (14 obras) · Artificial Intelligence (11 obras) · Philosophy (11 obras) · Mathematics (9 obras) · Phonetics and Phonology Research (9 obras) · Phonology (9 obras) · Artificial Intelligence (5 obras) · Natural language processing (5 obras)