Statistical Relational Learning With Unconventional String Models

Vu, Mai; Zehfroosh, Ashkan; Strother‐Garcia, Kristina; Sebok, Michael; Heinz, Jeffrey; Tanner, Herbert G.

doi:10.3389/frobt.2018.00076

Cited by 5 publications

(4 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Heinz and Rogers (2013) provide learning algorithms for the SL and SP classes as well as their Testable correlates. Other approaches have directly incorporated phonological features into the models (Vu et al 2018;. Learning of TSL classes has been discussed by Jardine and Heinz (2016) and Jardine and McMullin (2017), while online learners for this class and the remaining single-tier-based hierarchy were proposed by Lambert (2021).…”

Section: Further Readingmentioning

confidence: 99%

Typology emerges from simplicity in representations and learning

Lambert

Rawski

Heinz

2021

JLM

Self Cite

View full text Add to dashboard Cite

We derive well-understood and well-studied subregular classes of formal languages purely from the computational perspective of algorithmic learning problems. We parameterise the learning problem along dimensions of representation and inference strategy. Of special interest are those classes of languages whose learning algorithms are necessarily not prohibitively expensive in space and time, since learners are often exposed to adverse conditions and sparse data. Learned natural language patterns are expected to be most like the patterns in these classes, an expectation supported by previous typological and linguistic research in phonology. A second result is that the learning algorithms presented here are completely agnostic to choice of linguistic representation. In the case of the subregular classes, the results fall out from traditional model-theoretic treatments of words and strings. The same learning algorithms, however, can be applied to model-theoretic treatments of other linguistic representations such as syntactic trees or autosegmental graphs, which opens a useful direction for future research.

show abstract

Section: Further Readingmentioning

confidence: 99%

Typology emerges from simplicity in representations and learning

Lambert

Rawski

Heinz

2021

JLM

Self Cite

View full text Add to dashboard Cite

show abstract

“…In contrast, unconventional models for strings recognize that distinct alphabetic symbols may share properties, and expands the model signature by including these properties as unary relations (Strother- Garcia et al, 2016;Vu et al, 2018). For example, a conventional model of Σ = {a, .…”

Section: Unconventional Word Modelsmentioning

confidence: 99%

“…Different representations of strings and trees provide a unified perspective on well-known subclasses of the regular languages from a modeltheoretic and logical perspective (Thomas, 1997;Rogers et al, 2013). However, they also open up new doors for grammatical inference by allowing one to consider other models for strings (Strother-Garcia et al, 2016;Vu et al, 2018).…”

Section: Unconventional Word Modelsmentioning

confidence: 99%

Learning with Partially Ordered Representations

Chandlee¹,

Eyraud²,

Heinz³

et al. 2019

Proceedings of the 16th Meeting on the Mathematics of Language

Self Cite

View full text Add to dashboard Cite

This paper examines the characterization and learning of grammars defined with enriched representational models. Model-theoretic approaches to formal language theory traditionally assume that each position in a string belongs to exactly one unary relation. We consider unconventional string models where positions can have multiple, shared properties, which are arguably useful in many applications. We show the structures given by these models are partially ordered, and present a learning algorithm that exploits this ordering relation to effectively prune the hypothesis space. We prove this learning algorithm, which takes positive examples as input, finds the most general grammar which covers the data.

show abstract

“…Aprendizado de máquina relacional (AMR) destina-se à criação de modelos estatísticos para dados relacionais, isto é, dados cuja a informação relacional é tão ou mais importante que a informação individual de cada elemento. Essa classe de aprendizado tem sido utilizada em diversas aplicações, por exemplo, na extração de informação de dados não estruturados [Zhang et al 2016] e na modelagem de linguagem natural [Vu et al 2018]. Em particular, técnicas AMR têm sido amplamente empregadas em tarefas associadas a grafos de conhecimento, sobretudo na sua complementação [Nickel et al 2016].…”

Section: Aprendizado De Máquina Relacionalunclassified

Aprendizado de máquina e inferência em Grafos de Conhecimento

Silva¹,

Ziviani²,

Porto³

2019

Tópicos Em Gerenciamento De Dados E Informações: Minicursos Do SBBD 2019

View full text Add to dashboard Cite

The increasing production and availability of massive and heterogeneous data bring forward challenging opportunities. Among them, the development of computing systems capable of learning, reasoning, and inferring facts based on prior knowledge. In this scenario, knowledge bases are valuable assets for the knowledge representation and automated reasoning of diverse application domains. Especially, inference tasks on knowledge graphs (knowledge bases' graphical representations) are increasingly important in academia and industry. In this short course, we introduce machine learning methods and techniques employed in knowledge graph inference tasks as well as discuss the technical and scientific challenges and opportunities associated with those tasks. ResumoA crescente produção e disponibilização de dados caracterizados por heterogeneidade e larga escala apresentam oportunidades desafiadoras à nossa sociedade. Dentre elas, como construir sistemas computacionais capazes de aprender, raciocinar e realizar inferências sobre fatos a partir de conhecimento prévio é uma tarefa relevante. Nesse cenário, bases de conhecimento são ativos importantes na representação e raciocínio automatizado do conhecimento de diversos domínios de aplicação. Em especial, a inferência de informação a partir de sua representação em rede -grafos de conhecimentoganhou notoriedade na academia e indústria nos últimos anos. Em face ao exposto, neste curso, é apresentada uma introdução aos métodos e técnicas de aprendizado de máquina utilizadas em tarefas de inferência em grafos de conhecimento, discutindo-se os desafios e oportunidades tecnológicas e científicas desse tipo de tarefa. IntroduçãoA representação computacional de conhecimento remonta ao nascimento da área de Inteligência Artificial. Ela é motivada pela necessidade de que a informação sobre o mundo

show abstract

Statistical Relational Learning With Unconventional String Models

Cited by 5 publications

References 50 publications

Typology emerges from simplicity in representations and learning

Typology emerges from simplicity in representations and learning

Learning with Partially Ordered Representations

Aprendizado de máquina e inferência em Grafos de Conhecimento

Contact Info

Product

Resources

About