Skip to content

Commit

Permalink
Merge branch 'computgraph_pchou' of https://github.com/numericalEFT/F…
Browse files Browse the repository at this point in the history
…eynmanDiagram.jl into computgraph_pchou
  • Loading branch information
houpc committed Feb 10, 2024
2 parents a4d5c0c + d8b37c2 commit a188d68
Show file tree
Hide file tree
Showing 3 changed files with 3 additions and 10 deletions.
2 changes: 1 addition & 1 deletion .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -72,7 +72,7 @@ __pycache__
*.vscode

# Temporarily unignore vscode settings
!.vscode/
.vscode/

*.DS_Store
*.pb.gz
Expand Down
7 changes: 0 additions & 7 deletions .vscode/settings.json

This file was deleted.

4 changes: 2 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@
[![Build Status](https://github.com/numericalEFT/FeynmanDiagram.jl/workflows/CI/badge.svg)](https://github.com/numericalEFT/FeynmanDiagram.jl/actions)
[![Coverage](https://codecov.io/gh/numericalEFT/FeynmanDiagram.jl/branch/master/graph/badge.svg)](https://codecov.io/gh/numericalEFT/FeynmanDiagram.jl)

`FeynmanDiagram.jl` is a Julia package that introduces a novel compiler that transforms Feynman diagrams into a compact computational graph representation for efficient computation in Quantum Field Theory (QFT). It utilizes Taylor-mode Automatic Differentiation (AD) for field-theoretic renormalization, showcasing the synergy between QFT and AI tech stack to address computational challenges in QFT.
`FeynmanDiagram.jl` is a Julia package designed to efficiently encode Feynman diagrams --- essential elements of Quantum Field Theory (QFT) —-- into compact computational graphs for fast evaluation. It employs Taylor-mode Automatic Differentiation (AD) specifically to implement field-theoretic renormalization schemes, a pivotal technique in QFT that significantly improves the convergence of Feynman diagrammatic series. This approach underscores the synergy between QFT and AI technologies, effectively addressing the sophisticated computational challenges in QFT.

## Key Features

Expand All @@ -17,7 +17,7 @@

## Compiler Architecture Overview

In general, Feynman diagrams represents high-order integral. The integrand are propagators/interactions composed by the basis arithmetic operations (multiplication, addition). The sequence of calculating the integrand by combining the propagators/interactions with the arithmetic operatos can be represented as an algebraic computational graph. In this sense, the computational graph serves as an intermediate representation that standardizes the treatment of various diagram types, ensuring a consistent approach across different QFT calculations.
In general, Feynman diagrams represents high-order integral. The integrand are propagators/interactions composed by the basis arithmetic operations (multiplication, addition, power, etc). The sequence of calculating the integrand by combining the propagators/interactions with the arithmetic operatos can be represented as an algebraic computational graph. In this sense, the computational graph serves as an intermediate representation that standardizes the treatment of various diagram types, ensuring a consistent approach across different QFT calculations.

![infrastructure](assets/diagram_compiler.svg?raw=true "Compiler Infrastructure")

Expand Down

0 comments on commit a188d68

Please sign in to comment.