Summary of Learning Structure-aware Representations Of Dependent Types, by Konstantinos Kogkalidis et al.
Learning Structure-Aware Representations of Dependent Types
by Konstantinos Kogkalidis, Orestis Melkonian, Jean-Philippe Bernardy
First submitted to arxiv on: 3 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Programming Languages (cs.PL)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper introduces a novel dataset and neural architecture that combines machine learning with dependent type theory, specifically for the Agda programming language and proof assistant. The dataset contains elaborate program-proofs that can be used in various machine learning applications, allowing researchers to leverage the ultra-high resolution of proof states at the sub-type level. The proposed neural architecture aims to faithfully represent dependently-typed programs based on structural rather than nominal principles. Initial results show promising performance in a premise selection setup, surpassing strong baselines. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper brings together two important fields: machine learning and dependent type theory. It creates a special dataset that can be used for different machine learning tasks. The dataset has very detailed information about proof states at the smallest level. The paper also proposes a new way to build neural networks that can accurately represent programs with dependent types, which is important for building strong models. |
Keywords
* Artificial intelligence * Machine learning
