Shane Steinert-Threlkeld's repositories
decisions-games-ai
Lecture notes for a course on Decision and Game Theory for undergraduates studying AI
quantifier-rnn-learning
Accompanies the paper "Learnability and Semantic Universals" ; trains recurrent neural networks to learn to verify sentences with quantifiers in order to explain semantic universals
replicator_dynamic_examples
Simple code for running and visualizing replicator dynamics
color-learning
Accompanies the paper, "Ease of learning explains semantic universals". Generates and trains neural networks on artificial color systems, showing that _degree of convexity_ explains accuracy.
function-words-context
Signaling games with function words amidst contextual variability
nn-tutorial
A tutorial introduction to neural networks, with an eye towards linguistic applications.
responsive-verbs
Accompanies the paper, "An Explanation of the Veridical Uniformity Universal"; trains neural networks to learn so-called responsive verbs in order to explain a semantic universal.
OntologyCV
Files for a computational biology project to extract a cardiovascular ontology from Foundational Model of Anatomy and use it to label a left ventricle atlas and then locate regions of interest in experiments.
learning-evolution-compositionality
Companion to the paper: Shane Steinert-Threlkeld, "Compositional Signaling in a Complex World", Journal of Logic, Language, and Information, vol 25 no 3, pp. 379-397 (DOI: 10.1007/s10849-016-9236-9)
neural-vision-most
Accompanies the paper "Neural Models of the Psychosemantics of 'most'": trains neural models of visual attention to replicate varoius cognitive tasks
scale-free-protest
Code for a project on relation between scaling parameter in scale-free networks and size of protests
fill-in-the-quant
Predict the most suitable quantifier for a given linguistic context
function-word-signaling
Code and data for paper (to be) submitted to JoLLI
nlp-in-ling
Natural Language Processing Research in North American Linguistics Departments
obsidian-clipper-bookmarklet
A bookmarklet that saves the current page and/or highlighted text to Obsidian.md
python-tutorial-clms
Short Python tutorial for CLMS orientation
tensorflow
Computation using data flow graphs for scalable machine learning