# Learning Subjective Time-Series Data via Utopia Label Distribution Approximation

arXiv: https://arxiv.org/abs/2307.07682 · HTML: https://mlsanity.com/paper/2307.07682/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 62 references that are in the map.
- First posted on arXiv 15 Jul 2023.

## Builds on (most cited first, 20 of 20)

- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [Xception: Deep Learning with Depthwise Separable Convolutions](https://mlsanity.com/paper/1610.02357/) (2016)
- [OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields](https://mlsanity.com/paper/1812.08008/) (2018)
- [Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks](https://mlsanity.com/paper/1604.02878/) (2016)
- [Decoupling Representation and Classifier for Long-Tailed Recognition](https://mlsanity.com/paper/1910.09217/) (2019)
- [Large-Scale Long-Tailed Recognition in an Open World](https://mlsanity.com/paper/1904.05160/) (2019)
- [CNN Architectures for Large-Scale Audio Classification](https://mlsanity.com/paper/1609.09430/) (2016)
- [Rethinking the Value of Labels for Improving Class-Imbalanced Learning](https://mlsanity.com/paper/2006.07529/) (2020)
- [Delving into Deep Imbalanced Regression](https://mlsanity.com/paper/2102.09554/) (2021)
- [Balanced MSE for Imbalanced Visual Regression](https://mlsanity.com/paper/2203.16427/) (2022)
- [Video Summarization with Attention-Based Encoder-Decoder Networks](https://mlsanity.com/paper/1708.09545/) (2017)
- [RankSim: Ranking Similarity Regularization for Deep Imbalanced Regression](https://mlsanity.com/paper/2205.15236/) (2022)
- [Exploring global diverse attention via pairwise temporal relation for video summarization](https://mlsanity.com/paper/2009.10942/) (2020)
- [Supervised Video Summarization via Multiple Feature Sets with Parallel Attention](https://mlsanity.com/paper/2104.11530/) (2021)
- [Audio-Visual Fusion for Emotion Recognition in the Valence-Arousal Space Using Joint Cross-Attention](https://mlsanity.com/paper/2209.09068/) (2022)
- [Affect2MM: Affective Analysis of Multimedia Content Using Emotion Causality](https://mlsanity.com/paper/2103.06541/) (2021)
- [TAA-GCN: A Temporally Aware Adaptive Graph Convolutional Network for Age Estimation](https://mlsanity.com/paper/2305.08779/) (2023)
- [GLA in MediaEval 2018 Emotional Impact of Movies Task](https://mlsanity.com/paper/1911.12361/) (2019)

## Abstract (arXiv)

Subjective time-series regression (STR) tasks have gained increasing attention recently. However, most existing methods overlook the label distribution bias in STR data, which results in biased models. Emerging studies on imbalanced regression tasks, such as age estimation and depth estimation, hypothesize that the prior label distribution of the dataset is uniform. However, we observe that the label distributions of training and test sets in STR tasks are likely to be neither uniform nor identical. This distinct feature calls for new approaches that estimate more reasonable distributions to train a fair model. In this work, we propose Utopia Label Distribution Approximation (ULDA) for time-series data, which makes the training label distribution closer to real-world but unknown (utopia) label distribution. This would enhance the model's fairness. Specifically, ULDA first convolves the training label distribution by a Gaussian kernel. After convolution, the required sample quantity at each regression label may change. We further devise the Time-slice Normal Sampling (TNS) to generate new samples when the required sample quantity is greater than the initial sample quantity, and the Convolutional Weighted Loss (CWL) to lower the sample weight when the required sample quantity is less than the initial quantity. These two modules not only assist the model training on the approximated utopia label distribution, but also maintain the sample continuity in temporal context space. To the best of our knowledge, ULDA is the first method to address the label distribution bias in time-series data. Extensive experiments demonstrate that ULDA lifts the state-of-the-art performance on two STR tasks and three benchmark datasets.