Analyzing Correlated Evolution of Multiple Features Using Latent Representations

Murawaki, Yugo

doi:10.18653/v1/d18-1468

Cited by 7 publications

(4 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Considering the results of our reconstruction ( §S4, WO1-50), we have to bear in mind that our model does not take into account implicational dependencies between variables (e.g. head-final or head-initial; Pagel & Meade 2006, Dunn et al 2011, Murawaki 2018. The probability of values of a categorical variable is estimated independently of other variables.…”

Section: Verbal Morphology and Tensementioning

confidence: 99%

Reconstructing the evolution of Indo-European grammar

Carling¹,

Cathcart²

2021

Language

View full text Add to dashboard Cite

This study uses phylogenetic methods adopted from computational biology in order to reconstruct features of Proto-Indo-European morphosyntax. We estimate the probability of the presence of typological features in Proto-Indo-European on the assumption that these features change according to a stochastic process governed by evolutionary transition rates between them. We compare these probabilities to previous reconstructions of Proto-Indo-European morphosyntax, which use either the comparative-historical method or implicational typology. We find that our reconstruction yields strong support for a canonical model (synthetic, nominative-accusative, headfinal) of the protolanguage and low support for any alternative model. Observing the evolutionary dynamics of features in our data set, we conclude that morphological features have slower rates of change, whereas syntactic traits change faster. Additionally, more frequent, unmarked traits in grammatical hierarchies have slower change rates when compared to less frequent, marked ones, which indicates that universal patterns of economy and frequency impact language change within the family.

show abstract

Section: Verbal Morphology and Tensementioning

confidence: 99%

Reconstructing the evolution of Indo-European grammar

Carling¹,

Cathcart²

2021

Language

View full text Add to dashboard Cite

show abstract

“…For example, instead of a basic word order variable (BWO) as in The World Atlas of Language Structures (WALS, Dryer and Haspelmath ( 2013)) Feature 81A, the SSWL dataset has several surface word order variables such as SVO, SOV, etc. (For a discussion of deep and surface structure in word order features see Rizzi (2017) and also Murawaki (2018).) However, as demonstrated by previous analysis carried out on this data set (see for instance Port et al (2019); Ortegaray et al (2018)), the SSWL data still provide valid information regarding the distribution of syntactic features across world languages, and historical phenomena of syntactic relatedness.…”

Section: The Influence Of Sites In Leaf Sequencesmentioning

confidence: 75%

Syntactic Structures and the General Markov Models

Gakkhar,

Marcolli

2024

Math.Comput.Sci.

View full text Add to dashboard Cite

We further the theme of studying syntactic structures data from Longobardi (2017b), Collins (2010), Ceolin et al. (2020) and Koopman (2011) using general Markov models initiated in Shu et al. (2017), exploring the question of how consistent the data is with the idea that general Markov models. The ideas explored in the present paper are more generally applicable than to the setting of syntactic structures, and can be used when analyzing consistency of data with general Markov models. Additionally, we give an interpretation of the methods of Ceolin et al. (2020) as an infinite sites evolutionary model and compare it to the Markov model and explore each in the context of evolutionary processes acting on human language syntax.

show abstract

“…For language-level representations, URIEL and lang2vec ) allow a straightforward extraction of typological binary features from different KBs. Murawaki (2015Murawaki ( , 2017Murawaki ( , 2018 exploits them to build latent language representations with independent binary variables. Language features are encoded from data-driven tasks as well, such as NMT (Malaviya et al, 2017) or language modelling (Tsvetkov et al, 2016;Östling and Tiedemann, 2017;Bjerva and Augenstein, 2018b) with complementary linguistic-related target tasks (Bjerva and Augenstein, 2018a).…”

Section: Related Workmentioning

confidence: 99%

Bridging Linguistic Typology and Multilingual Machine Translation with Multi-View Language Representations

Oncevay

Haddow²,

Birch³

2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

Sparse language vectors from linguistic typology databases and learned embeddings from tasks like multilingual machine translation have been investigated in isolation, without analysing how they could benefit from each other's language characterisation. We propose to fuse both views using singular vector canonical correlation analysis and study what kind of information is induced from each source. By inferring typological features and language phylogenies, we observe that our representations embed typology and strengthen correlations with language relationships. We then take advantage of our multi-view language vector space for multilingual machine translation, where we achieve competitive overall translation accuracy in tasks that require information about language similarities, such as language clustering and ranking candidates for multilingual transfer. With our method, which is also released as a tool, we can easily project and assess new languages without expensive retraining of massive multilingual or ranking models, which are major disadvantages of related approaches.

show abstract

Analyzing Correlated Evolution of Multiple Features Using Latent Representations

Cited by 7 publications

References 38 publications

Reconstructing the evolution of Indo-European grammar

Reconstructing the evolution of Indo-European grammar

Syntactic Structures and the General Markov Models

Bridging Linguistic Typology and Multilingual Machine Translation with Multi-View Language Representations

Contact Info

Product

Resources

About