codes for Self-paced Deep Regression Forests with Consideration on Ranking Fairness

Related tags

Deep LearningSPUDRFs
Overview

Self-paced Deep Regression Forests with Consideration on Ranking Fairness

This is official codes for paper Self-paced Deep Regression Forests with Consideration on Ranking Fairness. In this paper, we proposes a new self-paced paradigm for deep discriminative model, which distinguishes noisy and underrepresented examples according to the output likelihood and entropy associated with each example, and we tackle the fundamental ranking problem in SPL from a new perspective: Fairness.

Why should we consider the fairness of self-paced learning?

We find that SPL focuses on easy samples at early pace and the underrepresented ones are always ranked at the end of the whole sequence. This phenomenon demonstrates the SPL has a potential sorting fairness issue. However, SPUDRFs considers sample uncertainty when ranking samples, thus making underrepresented samples be selected at early pace.

Tasks and Performances

Age Estimation on MORPH II Dataset

The gradual learning process of SP-DRFs and SPUDRFs. Left: The typical worst cases at each iteration. Right: The MAEs of SP-DRFs and SPUDRFs at each pace descend gradually. Compared with SP-DRFs, the SPUDRFs show its superiority of taking predictive uncertainty into consideration.

Gaze Estimation on MPII Dataset

The similar phenomena can be observed on MPII dataset.

Head Pose Estimation on BIWI Dataset

For visualization, we plot the leaf node distribution of SP-DRFs and SPUDRFs in gradual learning process. The means of leaf nodes of SP-DRFs gather in a small range, incurring seriously biased solutions, while that of SPUDRFs distribute widely, leading to much better MAE performance.

Fairness Evaluation

We use FRIA, proposed in our paper, as fairness metric. FAIR is defined as following form.

The following table shows the FAIR of different methods on different datasets. SPUDRFs achieve the best performance on all datasets.
Dataset MORPH FGNET BIWI BU-3DFE MPII
DRFs 0.46 0.42 0.46 0.740 0.67
SP-DRFs 0.44 0.37 0.43 0.72 0.67
SPUDRFs 0.48 0.42 0.70 0.76 0.69

How to train your SPUDRFs

Pre-trained models and Dataset

We use pre-trained models for our training. You can download VGGFace from here and VGG IMDB-WIKI from here. The datasets used in our experiment are in following table. We use MTCNN to detect and align face. For BIWI, we use depth images. For MPII, we use normalized left eye and right eye patch as input, and details about normalization can be found here.

Task Dataset
Age Estimation MOPRH and FG-NET
Head Estimation BIWI and BU-3DFE
Gaze Estimation MPII

Environment setup

All codes are based on Pytorch, before you run this repo, please make sure that you have a pytorch envirment. You can install them using following command.

pip install -r requirements.txt

Train SPUDRFs

Code descritption:

Here is the description of the main codes.

step.py:         train SPUDRFs from scratch  
train.py:        complete one pace training for a given train set
predict.py:      complete a test for a given test set
picksamples.py:  select samples for next pace   

Train your SPUDRFs from scratch:

You should download this repo, and prepare your datasets and pre-trained models, then just run following command to train your SPUDRFs from scratch.

  • Clone this repo:
git clone https://github.com/learninginvision/SPUDRFs.git  
cd SPUDFRs  
  • Set config.yml
lr: 0.00002
max_step: 80000
batchsize: 32

total_pace: 10
pace_percent: [0.5, 0.0556, 0.0556, 0.0556, 0.0556, 0.0556, 0.0556, 0.0556, 0.0556, 0.0552]
alpha: 2
threshold: -3.0
ent_pick_per: 0
capped: False
  • Train from scratch
python step.py

Acknowledgments

This code is inspired by caffe-DRFs.

Owner
Learning in Vision
Understanding and learning in computer vision.
Learning in Vision
Creating Multi Task Models With Keras

Creating Multi Task Models With Keras About The Project! I used the keras and Tensorflow Library, To build a Deep Learning Neural Network to Creating

Srajan Chourasia 4 Nov 28, 2022
DilatedNet in Keras for image segmentation

Keras implementation of DilatedNet for semantic segmentation A native Keras implementation of semantic segmentation according to Multi-Scale Context A

303 Mar 15, 2022
Learning to Segment Instances in Videos with Spatial Propagation Network

Learning to Segment Instances in Videos with Spatial Propagation Network This paper is available at the 2017 DAVIS Challenge website. Check our result

Jingchun Cheng 145 Sep 28, 2022
Fibonacci Method Gradient Descent

An implementation of the Fibonacci method for gradient descent, featuring a TKinter GUI for inputting the function / parameters to be examined and a matplotlib plot of the function and results.

Emma 1 Jan 28, 2022
Simulation code and tutorial for BBHnet training data

Simulation Dataset for BBHnet NOTE: OLD README, UPDATE IN PROGRESS We generate simulation dataset to train BBHnet, our deep learning framework for det

0 May 31, 2022
Implementing DropPath/StochasticDepth in PyTorch

%load_ext memory_profiler Implementing Stochastic Depth/Drop Path In PyTorch DropPath is available on glasses my computer vision library! Introduction

Francesco Saverio Zuppichini 13 Jan 05, 2023
Object-aware Contrastive Learning for Debiased Scene Representation

Object-aware Contrastive Learning Official PyTorch implementation of "Object-aware Contrastive Learning for Debiased Scene Representation" by Sangwoo

43 Dec 14, 2022
Code for WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models.

WECHSEL Code for WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models. arXiv: https://arx

Institute of Computational Perception 45 Dec 29, 2022
An algorithmic trading bot that learns and adapts to new data and evolving markets using Financial Python Programming and Machine Learning.

ALgorithmic_Trading_with_ML An algorithmic trading bot that learns and adapts to new data and evolving markets using Financial Python Programming and

1 Mar 14, 2022
This repository contains the implementation of the paper: "Towards Frequency-Based Explanation for Robust CNN"

RobustFreqCNN About This repository contains the implementation of the paper "Towards Frequency-Based Explanation for Robust CNN" arxiv. It primarly d

Sarosij Bose 2 Jan 23, 2022
DeepSTD: Mining Spatio-temporal Disturbances of Multiple Context Factors for Citywide Traffic Flow Prediction

DeepSTD: Mining Spatio-temporal Disturbances of Multiple Context Factors for Citywide Traffic Flow Prediction This is the implementation of DeepSTD in

5 Sep 26, 2022
Minecraft agent to farm resources using reinforcement learning

BarnyardBot CS 175 group project using Malmo download BarnyardBot.py into the python examples directory and run 'python BarnyardBot.py' in the console

0 Jul 26, 2022
Hierarchical Memory Matching Network for Video Object Segmentation (ICCV 2021)

Hierarchical Memory Matching Network for Video Object Segmentation Hongje Seong, Seoung Wug Oh, Joon-Young Lee, Seongwon Lee, Suhyeon Lee, Euntai Kim

Hongje Seong 72 Dec 14, 2022
An implementation of shampoo

shampoo.pytorch An implementation of shampoo, proposed in Shampoo : Preconditioned Stochastic Tensor Optimization by Vineet Gupta, Tomer Koren and Yor

Ryuichiro Hataya 69 Sep 10, 2022
Computational Pathology Toolbox developed by TIA Centre, University of Warwick.

TIA Toolbox Computational Pathology Toolbox developed at the TIA Centre Getting Started All Users This package is for those interested in digital path

Tissue Image Analytics (TIA) Centre 156 Jan 08, 2023
High-Resolution Image Synthesis with Latent Diffusion Models

Latent Diffusion Models arXiv | BibTeX High-Resolution Image Synthesis with Latent Diffusion Models Robin Rombach*, Andreas Blattmann*, Dominik Lorenz

CompVis Heidelberg 5.6k Dec 30, 2022
NeoPlay is the project dedicated to ESport events.

NeoPlay is the project dedicated to ESport events. On this platform users can participate in tournaments with prize pools as well as create their own tournaments.

3 Dec 18, 2021
End-to-End Speech Processing Toolkit

ESPnet: end-to-end speech processing toolkit system/pytorch ver. 1.3.1 1.4.0 1.5.1 1.6.0 1.7.1 1.8.1 1.9.0 ubuntu20/python3.9/pip ubuntu20/python3.8/p

ESPnet 5.9k Jan 04, 2023
Segmentation models with pretrained backbones. Keras and TensorFlow Keras.

Python library with Neural Networks for Image Segmentation based on Keras and TensorFlow. The main features of this library are: High level API (just

Pavel Yakubovskiy 4.2k Jan 09, 2023
Self-Correcting Quantum Many-Body Control using Reinforcement Learning with Tensor Networks

Self-Correcting Quantum Many-Body Control using Reinforcement Learning with Tensor Networks This repository contains the code and data for the corresp

Friederike Metz 7 Apr 23, 2022