[ICML 2021] Towards Understanding and Mitigating Social Biases in Language Models

Overview

Towards Understanding and Mitigating Social Biases in Language Models

This repo contains code and data for evaluating and mitigating bias from generation models.

Paper

Towards Understanding and Mitigating Social Biases in Language Models
Paul Pu Liang, Chiyu Wu, Louis-Philippe Morency, and Ruslan Salakhutdinov
ICML 2021

If you find this repository useful, please cite our paper:

@inproceedings{liang2021towards,
  title={Towards Understanding and Mitigating Social Biases in Language Models},
  author={Liang, Paul Pu and Wu, Chiyu and Morency, Louis-Philippe and Salakhutdinov, Ruslan},
  booktitle={International Conference on Machine Learning},
  pages={6565--6576},
  year={2021},
  organization={PMLR}
}

1. Identify bias-sensitive tokens, obtain bias subspace and create the dataset to train the bias classifier

python data_preprocess.py --embed_source glove --by_pca True --num_components 5 --save_subspace False

Glove embedding and gpt2 embedding are large files, you can download or extract them by yourself. We also provide the google drive link.

2. Train the bias classifier and learn the projection matrix P

python context_nullspace_projection.py

The code of nullspace projection is from INLP. Thanks for their great work!

To run the INLP experiments, you need to git clone https://github.com/shauli-ravfogel/nullspace_projection first, and put it under the root directory of this repo.

3. Evaluate Bias existing in the gpt2

Local Bias

cd src/local_bias
python measure_local_bias.py

It will take long time to run the evaluation script on the full data. Here we provide the subset of our evaluation data now. Full data will be uploaded via google drive soon.

Global Bias

We use the regard score difference as the metric for global bias. The evaluation code is from https://github.com/ewsheng/nlg-bias. Thanks for their great work!

git clone https://github.com/ewsheng/nlg-bias.git
cd src/global_bias
python generate_full_sentence.py --algorithm INLP

After full sentences are generated, you need to use the regard classifier to measure the global bias.

To reproduce the result in our paper, we also provide the projection matrix P for the gender bias test in data/saved_P/P_gender_test_79.npy

Acknowledgements

Owner
Paul Liang
Machine Learning Department, CMU
Paul Liang
Repository for MeshTalk supplemental material and code once the (already approved) 16 GHS captures our lab will make publicly available are released.

meshtalk This repository contains code to run MeshTalk for face animation from audio. If you use MeshTalk, please cite @inproceedings{richard2021mesht

Meta Research 221 Jan 06, 2023
CompilerGym is a library of easy to use and performant reinforcement learning environments for compiler tasks

CompilerGym is a library of easy to use and performant reinforcement learning environments for compiler tasks

Facebook Research 721 Jan 03, 2023
Code for the paper "Curriculum Dropout", ICCV 2017

Curriculum Dropout Dropout is a very effective way of regularizing neural networks. Stochastically "dropping out" units with a certain probability dis

Pietro Morerio 21 Jan 02, 2022
DiffQ performs differentiable quantization using pseudo quantization noise. It can automatically tune the number of bits used per weight or group of weights, in order to achieve a given trade-off between model size and accuracy.

Differentiable Model Compression via Pseudo Quantization Noise DiffQ performs differentiable quantization using pseudo quantization noise. It can auto

Facebook Research 145 Dec 30, 2022
FlingBot: The Unreasonable Effectiveness of Dynamic Manipulations for Cloth Unfolding

This repository contains code for training and evaluating FlingBot in both simulation and real-world settings on a dual-UR5 robot arm setup for Ubuntu 18.04

Columbia Artificial Intelligence and Robotics Lab 70 Dec 06, 2022
Outlier Exposure with Confidence Control for Out-of-Distribution Detection

OOD-detection-using-OECC This repository contains the essential code for the paper Outlier Exposure with Confidence Control for Out-of-Distribution De

Nazim Shaikh 64 Nov 02, 2022
Metrics to evaluate quality and efficacy of synthetic datasets.

An Open Source Project from the Data to AI Lab, at MIT Metrics for Synthetic Data Generation Projects Website: https://sdv.dev Documentation: https://

The Synthetic Data Vault Project 129 Jan 03, 2023
Learning with Subset Stacking

Learning with Subset Stacking (LESS) LESS is a new supervised learning algorithm that is based on training many local estimators on subsets of a given

S. Ilker Birbil 19 Oct 04, 2022
LBBA-boosted WSOD

LBBA-boosted WSOD Summary Our code is based on ruotianluo/pytorch-faster-rcnn and WSCDN Sincerely thanks for your resources. Newer version of our code

Martin Dong 20 Sep 19, 2022
The codes and related files to reproduce the results for Image Similarity Challenge Track 1.

ISC-Track1-Submission The codes and related files to reproduce the results for Image Similarity Challenge Track 1. Required dependencies To begin with

Wenhao Wang 115 Jan 02, 2023
Cerberus Transformer: Joint Semantic, Affordance and Attribute Parsing

Cerberus Transformer: Joint Semantic, Affordance and Attribute Parsing Paper Introduction Multi-task indoor scene understanding is widely considered a

62 Dec 05, 2022
Training deep models using anime, illustration images.

animeface deep models for anime images. Datasets anime-face-dataset Anime faces collected from Getchu.com. Based on Mckinsey666's dataset. 63.6K image

Tomoya Sawada 61 Dec 25, 2022
The codes for the work "Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation"

Swin-Unet The codes for the work "Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation"(https://arxiv.org/abs/2105.05537). A validatio

869 Jan 07, 2023
OMAMO: orthology-based model organism selection

OMAMO: orthology-based model organism selection OMAMO is a tool that suggests the best model organism to study a biological process based on orthologo

Dessimoz Lab 5 Apr 22, 2022
Code accompanying paper: Meta-Learning to Improve Pre-Training

Meta-Learning to Improve Pre-Training This folder contains code to run experiments in the paper Meta-Learning to Improve Pre-Training, NeurIPS 2021. P

28 Dec 31, 2022
Real-time ground filtering algorithm of cloud points acquired using Terrestrial Laser Scanner (TLS)

This repository contains tools to simulate the ground filtering process of a registered point cloud. The repository contains two filtering methods. The first method uses a normal vector, and fit to p

5 Aug 25, 2022
HEAM: High-Efficiency Approximate Multiplier Optimization for Deep Neural Networks

Approximate Multiplier by HEAM What's HEAM? HEAM is a general optimization method to generate high-efficiency approximate multipliers for specific app

4 Sep 11, 2022
Generating Anime Images by Implementing Deep Convolutional Generative Adversarial Networks paper

AnimeGAN - Deep Convolutional Generative Adverserial Network PyTorch implementation of DCGAN introduced in the paper: Unsupervised Representation Lear

Rohit Kukreja 23 Jul 21, 2022
novel deep learning research works with PaddlePaddle

Research 发布基于飞桨的前沿研究工作,包括CV、NLP、KG、STDM等领域的顶会论文和比赛冠军模型。 目录 计算机视觉(Computer Vision) 自然语言处理(Natrual Language Processing) 知识图谱(Knowledge Graph) 时空数据挖掘(Spa

1.5k Dec 29, 2022
《Single Image Reflection Removal Beyond Linearity》(CVPR 2019)

Single-Image-Reflection-Removal-Beyond-Linearity Paper Single Image Reflection Removal Beyond Linearity. Qiang Wen, Yinjie Tan, Jing Qin, Wenxi Liu, G

Qiang Wen 51 Jun 24, 2022