Skip to content
View JosephPotashnik's full-sized avatar
6 followers · 0 following

Achievements

Achievement: Pull Sharkx3Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: QuickdrawAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pull Sharkx3Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: QuickdrawAchievement: Arctic Code Vault Contributor

Block or report JosephPotashnik

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
josephpotashnik/README.md

Joseph Potashnik

I work on grammar induction, parsing, and computational models of linguistic structure. My current research studies how indirect negative evidence can guide greedy search over bounded context-free grammar hypothesis spaces.

I am an independent researcher and software engineer based in London, with a PhD in Linguistics from Tel Aviv University and a BSc in Physics and Computer Science. My background combines generative linguistics, formal language theory, machine learning, and high-performance research software.

Current Research

Greedy Grammar Induction with Indirect Negative Evidence

A computational linguistics and machine learning project on unsupervised grammar induction from positive evidence, using grammar-shape constraints and indirect negative evidence to guide search.

  • Paper: arXiv:2312.15321
  • Code: public reproducibility repository
  • Areas: grammar induction, computational linguistics, formal languages, parsing, unsupervised learning
  • Implementation: C#/.NET, parallel search, Earley/CKY-style parsing, grammar equivalence

Selected Projects

Earley Parser

A parser for context-free grammars, with a full-stack demonstration interface.

Density-Based Clustering

Implementation and visualization of Rodriguez and Laio’s density-peaks clustering algorithm.

PhD Dissertation

Scalarity and unaccusativity at the lexicon-syntax interface.

Background

  • PhD Linguistics, Tel Aviv University
  • MA Linguistics, summa cum laude
  • BSc Physics and Computer Science
  • Senior software engineering experience in high-performance industrial systems

Contact

Pinned Loading

  1. GreedyGrammarInduction-JMLR GreedyGrammarInduction-JMLR Public

    Greedy Grammar Induction with Indirect Negative Evidence

    C#

  2. EarleyParserREST EarleyParserREST Public

    RESTful API for Earley parsing algorithm - ASP.NET Core backend

    C#

  3. parser-ui parser-ui Public

    Interactive UI for Earley Parser - React/TypeScript frontend

    CSS

  4. EarleyParser EarleyParser Public

    An optimised Earley Parser

    HTML

  5. DensityPeaksClustering DensityPeaksClustering Public

    C# implementation of Rodriguez & Laio's Fast Clustering by Density Peaks algorithm

    C# 2 1

  6. density-peaks-ui density-peaks-ui Public

    FrontEnd (with React) For serving Density Peaks Clustering Algorithms

    JavaScript