This repo contains the source code for the website https://arranger1044.github.io/probabilistic-circuits/ which is a curated and reasoned list of papers on probabilistic circuits (PCs), computational graphs encoding tractable probability distributions.
All the material in this repo is released to the Public Domain. Feel free to clone, fork or complete and/or correct any of these lists.
To add, change or remove a paper on the website, please open a pull request!
This site harness Jekyll templates in github pages and their file-based model view. Each paper in the website is associated a markdown file under the _papers
folder. Modifications to the key-value pairs in this single file would be reflected to the whole website.
Mandatory keys in a paper description are:
layout
to be left topaper
ref
a string acting as a unique identifiertitle
the complete paper titledate
intended as a publication date (only the year matters)tags
a space-separated sequence of tags to classify the paper (see below)authors
a string with authors names, separated by commavenue
the publication venue (conference, journal name)
Optional keys are:
pdf
a link to a publicly readable version of the papercode
link to the code released with the paperabstract
the paper abstract, as a single stringbibtex
a string for the bibtex entry
The script dblp_to_md.py
is a quick and dirt way to generate a skeleton of a markdown file entry from the condensed bibtex as available from DBLP
Papers on PCs can be catalogued according to the following tags.
Models:
acs
: Arithmetic circuitscnets
: Cutset networksspns
: Sum-Product networksaogs
: And/Or graphspdgs
: Probabilistic decision graphspsdds
: Probabilistic sentential decision diagramspcs
: Other probabilistic circuits
Algorithms:
str-le
: Structure learningpar-le
: Parameter learningcomp
: Compilation
Inference:
mar
: Marginal inferencemap
: MAP inferencemmap
: Marginal MAP inferencediv
: Divergences, IPMsexp
: Expectationsmom
: Momentssam
: Samplingapp
: Approximate inferenceimp
: Imprecise probabilities
Applications:
cv
: Computer visionnlp
: Natural language processingseg
: Semantic segmentationact
: Activity recognitionspe
: Speech recognition and reconstructionrob
: Roboticsbio
: Computational biologythe
: Theoryppl
: Probabilistic Programmingrep
: Representation Learninghw
: Hardwaresw
: Softwarexai
: Explanationsmisc
: Other applications
Special thanks to Giuseppe Lobraico who taught me how to deal with the ruby stack behind Jekyll.