RankSim: Ranking Similarity Regularization for Deep Imbalanced Regression

Data imbalance, in which a plurality of the data samples come from a small proportion of labels, poses a challenge in training deep neural networks. Unlike classification, in regression the labels are continuous, potentially boundless, and form a natural ordering. These distinct features of regression call for new techniques that leverage the additional information encoded in label-space relationships. This paper presents the RankSim (ranking similarity) regularizer for deep imbalanced regression, which encodes an inductive bias that samples that are closer in label space should also be closer in feature space. In contrast to recent distribution smoothing based approaches, RankSim captures both nearby and distant relationships: for a given data sample, RankSim encourages the sorted list of its neighbors in label space to match the sorted list of its neighbors in feature space. RankSim is complementary to conventional imbalanced learning techniques, including re-weighting, two-stage training, and distribution smoothing, and lifts the state-of-the-art performance on three imbalanced regression benchmarks: IMDB-WIKI-DIR, AgeDB-DIR, and STS-B-DIR.

• Publication year

  2022

• Venue

  International Conference on Machine Learning

• Publication date

  2022-05-30

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.48550/arXiv.2205.15236arXiv 2205.15236
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Delving into Deep Imbalanced Regression

  2021influential reference
• Towards Calibrated Model for Long-Tailed Visual Recognition from Prior Perspective

  2021cited by this paper
• Density-based weighting for imbalanced regression

  2021cited by this paper
• Exploring Balanced Feature Spaces for Representation Learning

  2021cited by this paper
• Self-Damaging Contrastive Learning

  2021cited by this paper
• Image-Level or Object-Level? A Tale of Two Resampling Strategies for Long-Tailed Detection

  2021cited by this paper
• PML: Progressive Margin Loss for Long-tailed Age Classification

  2021cited by this paper
• Remix: Rebalanced Mixup

  2020cited by this paper
• Improving Calibration for Long-Tailed Recognition

  2020cited by this paper
• Posterior Re-calibration for Imbalanced Datasets

  2020cited by this paper
• Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effect

  2020cited by this paper
• Balanced Meta-Softmax for Long-Tailed Visual Recognition

  2020cited by this paper
• Rethinking the Value of Labels for Improving Class-Imbalanced Learning

  2020cited by this paper
• Deep Metric Learning to Rank

  2019cited by this paper
• SoDeep: A Sorting Deep Net to Learn Ranking Loss Surrogates

  2019cited by this paper
• Large-Scale Long-Tailed Recognition in an Open World

  2019cited by this paper
• Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss

  2019cited by this paper
• Learning With Average Precision: Training Image Retrieval With a Listwise Loss

  2019cited by this paper
• Implicit Semantic Data Augmentation for Deep Networks

  2019cited by this paper
• Decoupling Representation and Classifier for Long-Tailed Recognition

  2019influential reference
• Differentiable Ranks and Sorting using Optimal Transport

  2019cited by this paper
• Differentiation of Blackbox Combinatorial Solvers

  2019influential reference
• Optimizing Rank-Based Metrics With Blackbox Differentiation

  2019influential reference
• What is the Effect of Importance Weighting in Deep Learning?

  2018cited by this paper
• GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding

  2018influential reference
• Focal Loss for Dense Object Detection

  2017influential reference
• A systematic study of the class imbalance problem in convolutional neural networks

  2017cited by this paper
• Hashing as Tie-Aware Learning to Rank

  2017cited by this paper
• mixup: Beyond Empirical Risk Minimization

  2017cited by this paper
• SMOGN: a Pre-processing Approach for Imbalanced Regression

  2017cited by this paper
• SemEval-2017 Task 1: Semantic Textual Similarity Multilingual and Crosslingual Focused Evaluation

  2017cited by this paper
• AgeDB: The First Manually Collected, In-the-Wild Age Database

  2017cited by this paper
• Deep Expectation of Real and Apparent Age from a Single Image Without Facial Landmarks

  2016cited by this paper
• Learning Deep Embeddings with Histogram Loss

  2016cited by this paper
• Training Deep Neural Networks via Direct Loss Minimization

  2015cited by this paper
• SMOTE for Regression

  2013cited by this paper
• SoftRank: optimizing non-smooth rank metrics

  2008cited by this paper
• Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning

  2005cited by this paper
• SMOTE: Synthetic Minority Over-sampling Technique

  2002cited by this paper

• Highly Imbalanced Regression with Tabular Data in SEP and Other Applications

  2025cites this paper
• iBACon: imBalance-Aware Contrastive Learning for Time Series Forecasting

  2025cites this paper
• AI-driven Remote Facial Skin Hydration and TEWL Assessment from Selfie Images: A Systematic Solution

  2025cites this paper
• Balanced Sharpness-Aware Minimization for Imbalanced Regression

  2025influential citation
• Disentangled Deep Smoothed Bootstrap for Fair Imbalanced Regression

  2025cites this paper
• MemRank: Memory-Augmented Similarity Ranking for Video-Based Depression Severity Estimation

  2025cites this paper
• Improve Representation for Imbalanced Regression through Geometric Constraints

  2025influential citation
• AutoML‐Driven Soft Sensors for Real‐Time Monitoring of Amino Acids in Mammalian Perfusion Cultures

  2025cites this paper
• VOCAL: Visual Odometry via ContrAstive Learning

  2025cites this paper
• AdaPRL: Adaptive Pairwise Regression Learning with Uncertainty Estimation for Universal Regression Tasks

  2025influential citation
• ACCon: Angle-Compensated Contrastive Regularizer for Deep Regression

  2025influential citation
• Sample Selection via Contrastive Fragmentation for Noisy Label Regression

  2025influential citation
• A Study on Correlation Between WAT and CP Yields Based on Deep Learning Methods

  2025cites this paper
• PoGDiff: Product-of-Gaussians Diffusion Models for Imbalanced Text-to-Image Generation

  2025cites this paper
• Improving Deep Regression with Tightness

  2025influential citation
• A Comprehensive Survey on Imbalanced Data Learning

  2025cites this paper
• Regularization for Deep Imbalanced Regression Based on Quantitative Relationship

  2025cites this paper
• CLOC: Contrastive Learning for Ordinal Classification with Multi-Margin N-pair Loss

  2025cites this paper
• IRMAE-AKDE: A Novel Solution to Deep Imbalanced Regression for Performance Prediction of Rolled Steel Plate

  2025cites this paper
• Adaptive Multi-Scale Permutation Entropy Regularizer for Deep Imbalanced Fetal Brain Age Regression

  2025cites this paper
• Ensemble of Regressors to Handle Bias in Predicting Age from Facial Images

  2025cites this paper
• Dynamic Similarity Weighted Regression

  2025cites this paper
• SPECTRA: Spectral Target-Aware Graph Augmentation for Imbalanced Molecular Property Regression

  2025cites this paper
• Calibrated mixup for imbalanced regression on tabular data

  2025cites this paper
• Deep semi-supervised regression with cooperative uncertainty-consistency regularization and adaptive calibration

  2025cites this paper
• More Data or Better Algorithms: Latent Diffusion Augmentation for Deep Imbalanced Regression

  2025cites this paper
• Rank and Align: Towards Effective Source-free Graph Domain Adaptation

  2024cites this paper
• HOPE: Hybrid-Granularity Ordinal Prototype Learning for Progression Prediction of Mild Cognitive Impairment

  2024cites this paper
• Are we Making Much Progress? Revisiting Chemical Reaction Yield Prediction from an Imbalanced Regression Perspective

  2024cites this paper
• Gradient Aligned Regression via Pairwise Losses

  2024influential citation
• AFreeCA: Annotation-Free Counting for All

  2024cites this paper
• Boarding for ISS: Imbalanced Self-Supervised Discovery of a Scaled Autoencoder for Mixed Tabular Datasets

  2024cites this paper
• PP-Loss: An imbalanced regression loss based on plotting position for improved precipitation nowcasting

  2024cites this paper
• IM-Context: In-Context Learning for Imbalanced Regression Tasks

  2024cites this paper
• A temporal partial domain adaptation network for transferable prognostics across working conditions with insufficient data

  2024cites this paper
• Optimizing Automatic Speech Assessment: W-RankSim Regularization and Hybrid Feature Fusion Strategies

  2024cites this paper
• Interpretable Cascading Mixture-of-Experts for Urban Traffic Congestion Prediction

  2024cites this paper
• Deep Imbalanced Regression to Estimate Vascular Age from PPG Data: a Novel Digital Biomarker for Cardiovascular Health

  2024cites this paper
• Balanced Training for the End-to-End Autonomous Driving Model Based on Kernel Density Estimation

  2024cites this paper
• Dist Loss: Enhancing Regression in Few-Shot Region through Distribution Distance Constraint

  2024cites this paper
• Data Augmentation with Variational Autoencoder for Imbalanced Dataset

  2024cites this paper
• Revive Re-weighting in Imbalanced Learning by Density Ratio Estimation

  2024cites this paper
• I MPROVING D EEP R EGRESSION WITH O RDINAL E N - TROPY

  2023cites this paper
• Improving Deep Regression with Ordinal Entropy

  2023cites this paper
• AdaFlood: Adaptive Flood Regularization

  2023cites this paper
• Deep Imbalanced Regression via Hierarchical Classification Adjustment

  2023cites this paper
• Survey on Imbalanced Data, Representation Learning and SEP Forecasting

  2023cites this paper
• Noise-in, Bias-out: Balanced and Real-time MoCap Solving

  2023cites this paper
• ConR: Contrastive Regularizer for Deep Imbalanced Regression

  2023influential citation
• Toward a Balanced Feature Space for the Deep Imbalanced Regression

  2023influential citation
• A step towards understanding why classification helps regression

  2023cites this paper
• Generalized Oversampling for Learning from Imbalanced datasets and Associated Theory

  2023cites this paper
• Continual Learning in Predictive Autoscaling

  2023cites this paper
• Learning Subjective Time-Series Data via Utopia Label Distribution Approximation

  2023influential citation
• Learning-to-Rank Meets Language: Boosting Language-Driven Ordering Alignment for Ordinal Classification

  2023cites this paper
• Variational Imbalanced Regression

  2023influential citation
• Uncertainty Voting Ensemble for Imbalanced Deep Regression

  2023influential citation
• Semi-Supervised Graph Imbalanced Regression

  2023influential citation
• Ranking Regularization for Critical Rare Classes: Minimizing False Positives at a High True Positive Rate

  2023cites this paper
• Textless Speech-to-Music Retrieval Using Emotion Similarity

  2023cites this paper
• Rank-N-Contrast: Learning Continuous Representations for Regression

  2022cites this paper
• T OWARDS EXTRAPOLATION IN DEEP MATERIAL PROPERTY REGRESSION

  year unknowninfluential citation
• Smoothed Bootstrap et génération de données synthétiques pour la modélisation des extrêmes

  year unknowncites this paper
• Domain Generalizing DINO for Visual Regression via Latent Distractor Subspace Consistency

  year unknowncites this paper