# Inaccurate Label Distribution Learning

arXiv: https://arxiv.org/abs/2302.13000 · HTML: https://mlsanity.com/paper/2302.13000/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 67 references that are in the map.
- First posted on arXiv 25 Feb 2023.

## Builds on (most cited first, 11 of 11)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Convex optimization](https://mlsanity.com/paper/2106.01946/) (2021)
- [Label Distribution Learning](https://mlsanity.com/paper/1408.6027/) (2014)
- [Projection onto the probability simplex: An efficient algorithm with a simple proof, and an application](https://mlsanity.com/paper/1309.1541/) (2013)
- [SCUT-FBP5500: A Diverse Benchmark Dataset for Multi-Paradigm Facial Beauty Prediction](https://mlsanity.com/paper/1801.06345/) (2018)
- [Multi-View Spectral Clustering Tailored Tensor Low-Rank Representation](https://mlsanity.com/paper/2004.14705/) (2020)
- [Hybrid Routing Transformer for Zero-Shot Learning](https://mlsanity.com/paper/2203.15310/) (2022)
- [SCUT-FBP: A Benchmark Dataset for Facial Beauty Perception](https://mlsanity.com/paper/1511.02459/) (2015)
- [Towards Speaker Age Estimation with Label Distribution Learning](https://mlsanity.com/paper/2202.11424/) (2022)
- [ULTRA: Uncertainty-aware Label Distribution Learning for Breast Tumor Cellularity Assessment](https://mlsanity.com/paper/2206.06623/) (2022)
- [Ensemble Clustering via Co-association Matrix Self-enhancement](https://mlsanity.com/paper/2205.05937/) (2022)

## Built on by (most cited first, 10 of 10)

- [Class-wise Balancing Data Replay for Federated Class-Incremental Learning](https://mlsanity.com/paper/2507.07712/) (2025)
- [Label Distribution Learning with Biased Annotations by Learning Multi-Label Representation](https://mlsanity.com/paper/2502.01170/) (2025)
- [Can Class-Priors Help Single-Positive Multi-Label Learning?](https://mlsanity.com/paper/2309.13886/) (2023)
- [Inaccurate Label Distribution Learning with Dependency Noise](https://mlsanity.com/paper/2405.16474/) (2024)
- [Geometrically Regularized Transfer Learning with On-Manifold and Off-Manifold Perturbation](https://mlsanity.com/paper/2505.15191/) (2025)
- [Concentration Distribution Learning from Label Distributions](https://mlsanity.com/paper/2505.21576/) (2025)
- [Trustworthy Federated Label Distribution Learning under Annotation Quality Disparity](https://mlsanity.com/paper/2605.04827/) (2026)
- [Towards Better Performance in Incomplete LDL: Addressing Data Imbalance](https://mlsanity.com/paper/2410.13579/) (2024)
- [Disentangled Geometric Alignment with Adaptive Contrastive Perturbation for Reliable Domain Transfer](https://mlsanity.com/paper/2505.15241/) (2025)
- [Dynamic Epsilon Scheduling: A Multi-Factor Adaptive Perturbation Budget for Adversarial Training](https://mlsanity.com/paper/2506.04263/) (2025)

## Abstract (arXiv)

Label distribution learning (LDL) trains a model to predict the relevance of a set of labels (called label distribution (LD)) to an instance. The previous LDL methods all assumed the LDs of the training instances are accurate. However, annotating highly accurate LDs for training instances is time-consuming and very expensive, and in reality the collected LD is usually inaccurate and disturbed by annotating errors. For the first time, this paper investigates the problem of inaccurate LDL, i.e., developing an LDL model with noisy LDs. We assume that the noisy LD matrix is a linear combination of an ideal LD matrix and a sparse noise matrix. Consequently, the problem of inaccurate LDL becomes an inverse problem, where the objective is to recover the ideal LD and noise matrices from the noisy LDs. We hypothesize that the ideal LD matrix is low-rank due to the correlation of labels and utilize the local geometric structure of instances captured by a graph to assist in recovering the ideal LD. This is based on the premise that similar instances are likely to share the same LD. The proposed model is finally formulated as a graph-regularized low-rank and sparse decomposition problem and numerically solved by the alternating direction method of multipliers. Furthermore, a specialized objective function is utilized to induce a LD predictive model in LDL, taking into account the recovered label distributions. Extensive experiments conducted on multiple datasets from various real-world tasks effectively demonstrate the efficacy of the proposed approach. endabstract