ReLoss - Official implementation for paper "Relational Surrogate Loss Learning" ICLR 2022

Last update: Nov 22, 2022

Overview

Relational Surrogate Loss Learning (ReLoss)

Official implementation for paper "Relational Surrogate Loss Learning" in International Conference on Learning Representations (ICLR) 2022.

By Tao Huang, Zekang Li, Hua Lu, Yong Shan, Shusheng Yang, Yang Feng, Fei Wang, Shan You, Chang Xu.

Usage

Install ReLoss

pip install git+https://github.com/hunto/ReLoss.git

Or install for development:

git clone https://github.com/hunto/ReLoss
cd ReLoss
pip install -e .

Training with ReLoss

All the inputs and outputs of ReLoss are the same as the original loss.

classification

from reloss.cls import ReLoss
loss_fn = ReLoss()

human pose estimation

from reloss.pose import ReLoss
loss_fn = ReLoss(heatmap_size=(64, 48))

Citation

@inproceedings{
huang2022relational,
title={Relational Surrogate Loss Learning},
author={Tao Huang and Zekang Li and Hua Lu and Yong Shan and Shusheng Yang and Yang Feng and Fei Wang and Shan You and Chang Xu},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=dZPgfwaTaXv}
}

You might also like...

[ICLR 2022 Oral] F8Net: Fixed-Point 8-bit Only Multiplication for Network Quantization

F8Net Fixed-Point 8-bit Only Multiplication for Network Quantization (ICLR 2022 Oral) OpenReview | arXiv | PDF | Model Zoo | BibTex PyTorch implementa

76 Dec 13, 2022

[ICLR 2022] DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETR

DAB-DETR This is the official pytorch implementation of our ICLR 2022 paper DAB-DETR. Authors: Shilong Liu, Feng Li, Hao Zhang, Xiao Yang, Xianbiao Qi

336 Dec 25, 2022

Official code for the ICLR 2021 paper Neural ODE Processes

Neural ODE Processes Official code for the paper Neural ODE Processes (ICLR 2021). Abstract Neural Ordinary Differential Equations (NODEs) use a neura

50 Oct 28, 2022

Official repository for the ICLR 2021 paper Evaluating the Disentanglement of Deep Generative Models with Manifold Topology

Official repository for the ICLR 2021 paper Evaluating the Disentanglement of Deep Generative Models with Manifold Topology Sharon Zhou, Eric Zelikman

34 Nov 16, 2022

Official codebase for ICLR oral paper Unsupervised Vision-Language Grammar Induction with Shared Structure Modeling

CLIORA This is the official codebase for ICLR oral paper: Unsupervised Vision-Language Grammar Induction with Shared Structure Modeling. We introduce

32 Dec 23, 2022

The official implementation of NeMo: Neural Mesh Models of Contrastive Features for Robust 3D Pose Estimation [ICLR-2021]. https://arxiv.org/pdf/2101.12378.pdf

NeMo: Neural Mesh Models of Contrastive Features for Robust 3D Pose Estimation [ICLR-2021] Release Notes The offical PyTorch implementation of NeMo, p

76 Nov 23, 2022

Official implementation of Self-supervised Graph Attention Networks (SuperGAT), ICLR 2021.

SuperGAT Official implementation of Self-supervised Graph Attention Networks (SuperGAT). This model is presented at How to Find Your Friendly Neighbor

127 Dec 28, 2022

Official PyTorch implementation of "Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets" (ICLR 2021)

Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets This is the official PyTorch implementation for the paper Rapid Neural A

48 Dec 26, 2022

Official Implementation of 'UPDeT: Universal Multi-agent Reinforcement Learning via Policy Decoupling with Transformers' ICLR 2021(spotlight)

UPDeT Official Implementation of UPDeT: Universal Multi-agent Reinforcement Learning via Policy Decoupling with Transformers (ICLR 2021 spotlight) The

96 Dec 22, 2022

Comments

train_reloss.py is not runnable

Hi, Thanks for sharing your awesome work. I notice there are many typos and un-existed parameters in train_reloss.py, such as the code from line 33-40. 'i' is not defined, and 'logits, targets' from the corresponding batch are not used in the following code.

opened by lwmlyy 12
Update spearman.py

In https://arxiv.org/abs/2203.09630, we propose monotonic differentiable sorting networks which perform better than logistic with ART. Thus, we changed the default interpolation_type to cauchy in our recent update to diffsort. While cauchy might work better, I suggest specifying the interpolation type explicitly for reproducibility.

Best regards,

Felix

opened by Felix-Petersen 2

ReLoss - Official implementation for paper "Relational Surrogate Loss Learning" ICLR 2022

Related tags

Overview

Relational Surrogate Loss Learning (ReLoss)

Usage

Install ReLoss

Training with ReLoss

Citation

You might also like...

[ICLR 2022 Oral] F8Net: Fixed-Point 8-bit Only Multiplication for Network Quantization

[ICLR 2022] DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETR

Official code for the ICLR 2021 paper Neural ODE Processes

Official repository for the ICLR 2021 paper Evaluating the Disentanglement of Deep Generative Models with Manifold Topology

Official codebase for ICLR oral paper Unsupervised Vision-Language Grammar Induction with Shared Structure Modeling

The official implementation of NeMo: Neural Mesh Models of Contrastive Features for Robust 3D Pose Estimation [ICLR-2021]. https://arxiv.org/pdf/2101.12378.pdf

Official implementation of Self-supervised Graph Attention Networks (SuperGAT), ICLR 2021.

Official PyTorch implementation of "Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets" (ICLR 2021)

Official Implementation of 'UPDeT: Universal Multi-agent Reinforcement Learning via Policy Decoupling with Transformers' ICLR 2021(spotlight)

Comments

train_reloss.py is not runnable

Update spearman.py

Releases(v1.0.0)

v1.0.0(Feb 18, 2022)

Owner

Tao Huang

Numba-accelerated Pythonic implementation of MPDATA with examples in Python, Julia and Matlab

Companion code for "Bayesian logistic regression for online recalibration and revision of risk prediction models with performance guarantees"

The code for Bi-Mix: Bidirectional Mixing for Domain Adaptive Nighttime Semantic Segmentation

Bayes-Newton—A Gaussian process library in JAX, with a unifying view of approximate Bayesian inference as variants of Newton's algorithm.

一个多语言支持、易使用的 OCR 项目。An easy-to-use OCR project with multilingual support.

A PyTorch implementation of "SelfGNN: Self-supervised Graph Neural Networks without explicit negative sampling"

[CVPR 2022] Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels

Distributed Asynchronous Hyperparameter Optimization better than HyperOpt.

Using Self-Supervised Pretext Tasks for Active Learning - Official Pytorch Implementation

Reviving Iterative Training with Mask Guidance for Interactive Segmentation

A Kaggle competition: discriminate gender based on handwriting

Code for Emergent Translation in Multi-Agent Communication

This repository contains datasets and baselines for benchmarking Chinese text recognition.

Group Activity Recognition with Clustered Spatial Temporal Transformer

KIND: an Italian Multi-Domain Dataset for Named Entity Recognition

The official implementation of the CVPR2021 paper: Decoupled Dynamic Filter Networks

codebase for "A Theory of the Inductive Bias and Generalization of Kernel Regression and Wide Neural Networks"

Python Environment for Bayesian Learning

Discord Multi Tool that focuses on design and easy usage

Source code for paper "Deep Superpixel-based Network for Blind Image Quality Assessment"