Decoupled Smoothing in Probabilistic Soft Logic

Overview

Decoupled Smoothing in Probabilistic Soft Logic

Experiments for "Decoupled Smoothing in Probabilistic Soft Logic".

Probabilistic Soft Logic

Probabilistic Soft Logic (PSL) is a machine learning framework for developing probabilistic models. You can find more information about PSL available at the PSL homepage and examples of PSL.

Documentation

This repository contains code to run PSL rules for one-hop method, two-hop method, and decoupled smoothing method for predicting genders in a social network. We provide links to the datasets (Facebook100) in the data sub-folder.

Obtaining the data

This repository set-up assumes that the FB100 (raw .mat files) have been acquired and are saved the data folder. Follow these steps:

  1. The Facebook100 (FB100) dataset is publicly available from the Internet Archive at https://archive.org/details/oxford-2005-facebook-matrix and other public repositories. Download the datasets.
  2. Save raw datasets in placeholder folder data. They should be in the following form: Amherst41.mat.

Set permissions

Make sure that permissions are set so you can run the run scripts:

chmod -R +x *

Reproducing results

Step 1: Generate input files

To reproduce the results, first need to generate the predicate txts, run ./generate_data.sh {school name}. It will automatically generate the files required to run the PSL models as well as the files to run the baseline model.

For example, to generate data using Amherst college as dataset, simply run ./generate_data.sh Amherst41.

Step 2: Run PSL models

Simple Exeucution

To reproduce the results of a specific PSL model, run ./run_all.sh {data} {method dir}. This will run a selected method for all random seeds at all percentages.

This takes the following positional parameters:

  • data: what datafile you would like to use
  • method dir: this is the path to the directory you'd like the run

For example, to reproduce the result for method one-hop using the Amherst college as dataset, simply run ./run_all.sh Amherst41 cli_one_hop.

Advanced Execution

If you need to get results for a more specific setting, run ./run_method.sh {data} {random seed} {precent labeled} {eval|learn} {method dir}. It runs a selected method for a specified seed for a specified percentage for either learning or evaluation.

This takes the following positional parameters:

  • data: what datafile you would like to use
  • random seed: what seed to use
  • percent labeled: what percentage of labeled data
  • {learn|eval}: specify if you're learning or evaluating
  • method dir: this is the path to the directory you'd like the run

The output will be written in the following directory: ../results/decoupled-smoothing/{eval|learn}/{method run}/{data used}/{random seed}/

The directory will contain a set of folders for the inferences found at each percent labeled, named inferred-predicates{pct labeled}. The folder will also contain the a copy of the base.data, gender.psl, files and output logs from the runs.

Step 3: Run baseline Decoupled Smoothing model

To run the baseline decoupled smoothing model, run baseline_ds.py. It will generate a csv file contains the results of the baseline model named baseline_result.csv.

Evaluation

To run the evaluation of each models, run evaluation.py, which will generate the two plots in Figure 3 in the paper.

Requirements

These experiments expect that you are running on a POSIX (Linux/Mac) system. The specific application dependencies are as follows:

  • Python3
  • Bash >= 4.0
  • PostgreSQL >= 9.5
  • Java >= 7

Citation

All of these experiments are discussed in the following paper:

@inproceedings{chen:mlg20,
    title = {Decoupled Smoothing in Probabilistic Soft Logic},
    author = {Yatong Chen and Byran Tor and Eriq Augustine and Lise Getoor},
    booktitle = {International Workshop on Mining and Learning with Graphs (MLG)},
    year = {2020},
    publisher = {MLG},
    address = {Virtual},
}
You might also like...
The official implementation of the CVPR2021 paper: Decoupled Dynamic Filter Networks
The official implementation of the CVPR2021 paper: Decoupled Dynamic Filter Networks

Decoupled Dynamic Filter Networks This repo is the official implementation of CVPR2021 paper: "Decoupled Dynamic Filter Networks". Introduction DDF is

Implement Decoupled Neural Interfaces using Synthetic Gradients in Pytorch
Implement Decoupled Neural Interfaces using Synthetic Gradients in Pytorch

disclaimer: this code is modified from pytorch-tutorial Image classification with synthetic gradient in Pytorch I implement the Decoupled Neural Inter

An implementation for the loss function proposed in Decoupled Contrastive Loss paper.

Decoupled-Contrastive-Learning This repository is an implementation for the loss function proposed in Decoupled Contrastive Loss paper. Requirements P

Code for paper Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic Forecasting
Code for paper Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic Forecasting

Decoupled Spatial-Temporal Graph Neural Networks Code for our paper: Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic Forecasting.

Aircache is an open-source caching and security solution that can be integrated with most decoupled apps that use REST APIs for communicating.

AirCache Aircache is an open-source caching and security solution that can be integrated with most decoupled apps that use REST APIs for communicating

LVI-SAM: Tightly-coupled Lidar-Visual-Inertial Odometry via Smoothing and Mapping
LVI-SAM: Tightly-coupled Lidar-Visual-Inertial Odometry via Smoothing and Mapping

LVI-SAM This repository contains code for a lidar-visual-inertial odometry and mapping system, which combines the advantages of LIO-SAM and Vins-Mono

Minimal implementation of Denoised Smoothing: A Provable Defense for Pretrained Classifiers in TensorFlow.
Minimal implementation of Denoised Smoothing: A Provable Defense for Pretrained Classifiers in TensorFlow.

Denoised-Smoothing-TF Minimal implementation of Denoised Smoothing: A Provable Defense for Pretrained Classifiers in TensorFlow. Denoised Smoothing is

Implementation of Online Label Smoothing in PyTorch

Online Label Smoothing Pytorch implementation of Online Label Smoothing (OLS) presented in Delving Deep into Label Smoothing. Introduction As the abst

Tensorflow Implementation of SMU: SMOOTH ACTIVATION FUNCTION FOR DEEP NETWORKS USING SMOOTHING MAXIMUM TECHNIQUE

SMU A Tensorflow Implementation of SMU: SMOOTH ACTIVATION FUNCTION FOR DEEP NETWORKS USING SMOOTHING MAXIMUM TECHNIQUE arXiv https://arxiv.org/abs/211

Image Processing, Image Smoothing, Edge Detection and Transforms

opevcvdl-hw1 This project uses openCV and Qt to achieve the requirements. Version Python 3.7 opencv-contrib-python 3.4.2.17 Matplotlib 3.1.1 pyqt5 5.1

Supervised Sliding Window Smoothing Loss Function Based on MS-TCN for Video Segmentation

SSWS-loss_function_based_on_MS-TCN Supervised Sliding Window Smoothing Loss Function Based on MS-TCN for Video Segmentation Supervised Sliding Window

Image Smoothing and Blurring Using OpenCV

Image-Smoothing-and-Blurring-Using-OpenCV This repository contains codes for performing image smoothing and blurring using OpenCV. There are different

Bucatini: a soft PIPE PHY for FPGA SerDes

Bucatini: a soft PIPE PHY for FPGA SerDes Bucatini is a noodly gateware layer capable of transforming an FPGA SerDes into a PIPE PHY, allowing you to

Custom TensorFlow2 implementations of forward and backward computation of soft-DTW algorithm in batch mode.

Batch Soft-DTW(Dynamic Time Warping) in TensorFlow2 including forward and backward computation Custom TensorFlow2 implementations of forward and backw

PyTorch implementation of Soft-DTW: a Differentiable Loss Function for Time-Series in CUDA
PyTorch implementation of Soft-DTW: a Differentiable Loss Function for Time-Series in CUDA

Soft DTW Loss Function for PyTorch in CUDA This is a Pytorch Implementation of Soft-DTW: a Differentiable Loss Function for Time-Series which is batch

Official implementation of "DSP: Dual Soft-Paste for Unsupervised Domain Adaptive Semantic Segmentation"

DSP Official implementation of "DSP: Dual Soft-Paste for Unsupervised Domain Adaptive Semantic Segmentation". Accepted by ACM Multimedia 2021. Authors

pytorch implementation of
pytorch implementation of "Distilling a Neural Network Into a Soft Decision Tree"

Soft-Decision-Tree Soft-Decision-Tree is the pytorch implementation of Distilling a Neural Network Into a Soft Decision Tree, paper recently published

 Fast Soft Color Segmentation
Fast Soft Color Segmentation

Fast Soft Color Segmentation

Permute Me Softly: Learning Soft Permutations for Graph Representations

Permute Me Softly: Learning Soft Permutations for Graph Representations

Owner
Kushal Shingote
Android Developer📱📱 iOS Apps📱📱 Swift | Xcode | SwiftUI iOS Swift development📱 Kotlin Application📱📱 iOS📱 Artificial Intelligence 💻 Data science
Kushal Shingote
VHDL to Discrete Logic on PCB Flow

PCBFlow Highly experimental set of scripts to transform a digital circuit described in a hardware description language (VHDL or Verilog) into a discre

Tim 77 Nov 4, 2022
ChronoRace is a tool to accurately perform timed race conditions to circumvent application business logic.

ChronoRace is a tool to accurately perform timed race conditions to circumvent application business logic. I've found in my research that w

Tanner 64 Aug 4, 2022
This is a Saleae Logic custom high level analyzer that allows you to search and mark specific packets.

SaleaePacketParser This is a Saleae Logic custom high level analyzer that allows you to search and mark specific packets. Field "Search For" is used f

null 1 Dec 16, 2021
VacationCycleLogicBackEnd - Vacation Cycle Logic BackEnd With Python

Vacation Cycle Logic BackEnd Getting Started Existing virtualenv If your project

Mohamed Gamal 0 Jan 3, 2022
Release for Improved Denoising Diffusion Probabilistic Models

improved-diffusion This is the codebase for Improved Denoising Diffusion Probabilistic Models. Usage This section of the README walks through how to t

OpenAI 1.2k Dec 30, 2022
A python package to adjust the bias of probabilistic forecasts/hindcasts using "Mean and Variance Adjustment" method.

Documentation A python package to adjust the bias of probabilistic forecasts/hindcasts using "Mean and Variance Adjustment" method. Read documentation

null 1 Feb 2, 2022
Logic-Sim - A clone of 'Digital Logic Sim' from Sebastian Lague

Logic Simulator This is a clone of 'Digital Logic Sim' from Sebastian Lague. But

Ethan 1 Feb 1, 2022
DeepProbLog is an extension of ProbLog that integrates Probabilistic Logic Programming with deep learning by introducing the neural predicate.

DeepProbLog DeepProbLog is an extension of ProbLog that integrates Probabilistic Logic Programming with deep learning by introducing the neural predic

KU Leuven Machine Learning Research Group 94 Dec 18, 2022
Probabilistic Programming in Python: Bayesian Modeling and Probabilistic Machine Learning with Theano

PyMC3 is a Python package for Bayesian statistical modeling and Probabilistic Machine Learning focusing on advanced Markov chain Monte Carlo (MCMC) an

PyMC 7.2k Dec 30, 2022
Probabilistic Programming in Python: Bayesian Modeling and Probabilistic Machine Learning with Aesara

PyMC3 is a Python package for Bayesian statistical modeling and Probabilistic Machine Learning focusing on advanced Markov chain Monte Carlo (MCMC) an

PyMC 7.2k Dec 30, 2022