Towards Debiasing NLU Models from Unknown Biases

Overview

Towards Debiasing NLU Models from Unknown Biases

Abstract: NLU models often exploit biased features to achieve high dataset-specific performance without properly learning the intended task. Recently proposed debiasing methods are shown to be effective in mitigating this tendency. However, these methods rely on a major assumption that the type of biased features is known a-priori, which limits their application to many NLU tasks and datasets. In this work, we present the first step to bridge this gap by introducing a self-debiasing framework that prevents models from mainly utilizing biases without knowing them in advance. The proposed framework is general and complementary to the existing debiasing methods. We show that the proposed framework allows these existing methods to retain the improvement on the challenge datasets (i.e., sets of examples designed to expose models’ reliance to biases) without specifically targeting certain biases. Furthermore, the evaluation suggests that applying the framework results in improved overall robustness.

The repository contains the code to reproduce our work in debiasing NLU models without prior information on biases. We provide 3 runs of experiment that are shown in our paper:

  1. Debias MNLI model from syntactic bias and evaluate on HANS as the out-of-distribution data using example reweighting.
  2. Debias MNLI model from syntactic bias and evaluate on HANS as the out-of-distribution data using product of expert.
  3. Debias MNLI model from syntactic bias and evaluate on HANS as the out-of-distribution data using confidence regularization.

Requirements

The code requires python >= 3.6 and pytorch >= 1.1.0.

Additional required dependencies can be found in requirements.txt. Install all requirements by running:

pip install -r requirements.txt

Data

Our experiments use MNLI dataset version provided by GLUE benchmark. Download the file from here, and unzip under the directory ./dataset The dataset directory should be structured as the following:

└── dataset 
    └── MNLI
        ├── train.tsv
        ├── dev_matched.tsv
        ├── dev_mismatched.tsv
        ├── dev_mismatched.tsv

Running the experiments

For each evaluation setting, use the --mode arguments to set the appropriate loss function. Choose the annealed version of the loss function for reproducing the annealed results.

To reproduce our result on MNLI ⮕ HANS, run the following:

cd src/
CUDA_VISIBLE_DEVICES=9 python train_distill_bert.py \
  --output_dir ../experiments_self_debias_mnli_seed111/bert_reweighted_sampled2K_teacher_seed111_annealed_1to08 \
  --do_train --do_eval --mode reweight_by_teacher_annealed \
  --custom_teacher ../teacher_preds/mnli_trained_on_sample2K_seed111.json --seed 111 --which_bias hans

Biased examples identification

To obtain predictions of the shallow models, we train the same model architecture on the fraction of the dataset. For MNLI we subsample 2000 examples and train the model for 5 epochs. For obtaining shallow models of other datasets please see the appendix of our paper. The shallow model can be obtained with the command below:

cd src/
CUDA_VISIBLE_DEVICES=9 python train_distill_bert.py \
 --output_dir ../experiments_shallow_mnli/bert_base_sampled2K_seed111 \
 --do_train --do_eval --do_eval_on_train --mode none\
 --seed 111 --which_bias hans --debug --num_train_epochs 5 --debug_num 2000

Once the training and the evaluation on train set is done, copy the probability json files in the output directory to ../teacher_preds/mnli_trained_on_sample2K_seed111.json.

Expected results

Results on the MNLI ⮕ HANS setting without annealing:

Mode Seed MNLI-m MNLI-mm HANS avg.
None 111 84.57 84.72 62.04
reweighting 111 81.8 82.3 72.1
PoE 111 81.5 81.1 70.3
conf-reg 222 83.7 84.1 68.7
Owner
Ubiquitous Knowledge Processing Lab
Ubiquitous Knowledge Processing Lab
ByteTrack超详细教程!训练自己的数据集&&摄像头实时检测跟踪

ByteTrack超详细教程!训练自己的数据集&&摄像头实时检测跟踪

Double-zh 45 Dec 19, 2022
"MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstruction" (CVPRW 2022) & (Winner of NTIRE 2022 Challenge on Spectral Reconstruction from RGB)

MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstruction (CVPRW 2022) Yuanhao Cai, Jing Lin, Zudi Lin, Haoqian Wang, Yulun Z

Yuanhao Cai 274 Jan 05, 2023
Open source repository for the code accompanying the paper 'PatchNets: Patch-Based Generalizable Deep Implicit 3D Shape Representations'.

PatchNets This is the official repository for the project "PatchNets: Patch-Based Generalizable Deep Implicit 3D Shape Representations". For details,

16 May 22, 2022
NHS AI Lab Skunkworks project: Long Stayer Risk Stratification

NHS AI Lab Skunkworks project: Long Stayer Risk Stratification A pilot project for the NHS AI Lab Skunkworks team, Long Stayer Risk Stratification use

NHSX 21 Nov 14, 2022
Good Semi-Supervised Learning That Requires a Bad GAN

Good Semi-Supervised Learning that Requires a Bad GAN This is the code we used in our paper Good Semi-supervised Learning that Requires a Bad GAN Ziha

Zhilin Yang 177 Dec 12, 2022
Contrastive Learning for Metagenomic Binning

CLMB A simple framework for CLMB - a novel deep Contrastive Learningfor Metagenomic Binning Created by Pengfei Zhang, senior of Department of Computer

1 Sep 14, 2022
😇A pyTorch implementation of the DeepMoji model: state-of-the-art deep learning model for analyzing sentiment, emotion, sarcasm etc

------ Update September 2018 ------ It's been a year since TorchMoji and DeepMoji were released. We're trying to understand how it's being used such t

Hugging Face 865 Dec 24, 2022
Paper: De-rendering Stylized Texts

Paper: De-rendering Stylized Texts Wataru Shimoda1, Daichi Haraguchi2, Seiichi Uchida2, Kota Yamaguchi1 1CyberAgent.Inc, 2 Kyushu University Accepted

CyberAgent AI Lab 55 Dec 18, 2022
CAPITAL: Optimal Subgroup Identification via Constrained Policy Tree Search

CAPITAL: Optimal Subgroup Identification via Constrained Policy Tree Search This repository is the official implementation of CAPITAL: Optimal Subgrou

Hengrui Cai 0 Oct 19, 2021
Car Price Predictor App used to predict the price of the car based on certain input parameters created using python's scikit-learn, fastapi, numpy and joblib packages.

Pricefy Car Price Predictor App used to predict the price of the car based on certain input parameters created using python's scikit-learn, fastapi, n

Siva Prakash 1 May 10, 2022
ROS-UGV-Control-Interface - Control interface which can be used in any UGV

ROS-UGV-Control-Interface Cam Closed: Cam Opened:

Ahmet Fatih Akcan 1 Nov 04, 2022
Learning to trade under the reinforcement learning framework

Trading Using Q-Learning In this project, I will present an adaptive learning model to trade a single stock under the reinforcement learning framework

Uirá Caiado 470 Nov 28, 2022
OstrichRL: A Musculoskeletal Ostrich Simulation to Study Bio-mechanical Locomotion.

OstrichRL This is the repository accompanying the paper OstrichRL: A Musculoskeletal Ostrich Simulation to Study Bio-mechanical Locomotion. It contain

Vittorio La Barbera 51 Nov 17, 2022
Async API for controlling Hue Lights

Hue API Async API for controlling Hue Lights Documentation: hue-api.nirantak.com Source: github.com/nirantak/hue-api Installation This is an async cli

Nirantak Raghav 4 Nov 16, 2022
Another pytorch implementation of FCN (Fully Convolutional Networks)

FCN-pytorch-easiest Trying to be the easiest FCN pytorch implementation and just in a get and use fashion Here I use a handbag semantic segmentation f

Y. Dong 158 Dec 21, 2022
Official implementation for "Low-light Image Enhancement via Breaking Down the Darkness"

Low-light Image Enhancement via Breaking Down the Darkness by Qiming Hu, Xiaojie Guo. 1. Dependencies Python3 PyTorch=1.0 OpenCV-Python, TensorboardX

Qiming Hu 30 Jan 01, 2023
Official implementation of "One-Shot Voice Conversion with Weight Adaptive Instance Normalization".

One-Shot Voice Conversion with Weight Adaptive Instance Normalization By Shengjie Huang, Yanyan Xu*, Dengfeng Ke*, Mingjie Chen, Thomas Hain. This rep

31 Dec 07, 2022
Single object tracking and segmentation.

Single/Multiple Object Tracking and Segmentation Codes and comparison of recent single/multiple object tracking and segmentation. News 💥 AutoMatch is

ZP ZHANG 385 Jan 02, 2023
This is the source code of the 1st place solution for segmentation task (with Dice 90.32%) in 2021 CCF BDCI challenge.

1st place solution in CCF BDCI 2021 ULSEG challenge This is the source code of the 1st place solution for ultrasound image angioma segmentation task (

Chenxu Peng 30 Nov 22, 2022
A C implementation for creating 2D voronoi diagrams

Branch OSX/Linux Windows master dev jc_voronoi A fast C/C++ header only implementation for creating 2D Voronoi diagrams from a point set Uses Fortune'

Mathias Westerdahl 481 Dec 29, 2022