# Alleviating Performance Disparity in Adversarial Spatiotemporal Graph Learning Under Zero-Inflated Distribution

arXiv: https://arxiv.org/abs/2504.00721 · HTML: https://mlsanity.com/paper/2504.00721/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 54 references that are in the map.
- First posted on arXiv 1 Apr 2025.

## Builds on (most cited first, 24 of 24)

- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [Graph WaveNet for Deep Spatial-Temporal Graph Modeling](https://mlsanity.com/paper/1906.00121/) (2019)
- [Adversarial Risk and the Dangers of Evaluating Against Weak Attacks](https://mlsanity.com/paper/1802.05666/) (2018)
- [Balanced Contrastive Learning for Long-Tailed Visual Recognition](https://mlsanity.com/paper/2207.09052/) (2022)
- [Variational Autoencoder for Deep Learning of Images, Labels and Captions](https://mlsanity.com/paper/1609.08976/) (2016)
- [To be Robust or to be Fair: Towards Fairness in Adversarial Training](https://mlsanity.com/paper/2010.06121/) (2020)
- [Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning](https://mlsanity.com/paper/2106.02584/) (2021)
- [Equalization Loss v2: A New Gradient Balance Approach for Long-tailed Object Detection](https://mlsanity.com/paper/2012.08548/) (2020)
- [Balanced MSE for Imbalanced Visual Regression](https://mlsanity.com/paper/2203.16427/) (2022)
- [Rank-N-Contrast: Learning Continuous Representations for Regression](https://mlsanity.com/paper/2210.01189/) (2022)
- [Hard Sample Aware Network for Contrastive Deep Graph Clustering](https://mlsanity.com/paper/2212.08665/) (2022)
- [Adversarial Robustness under Long-Tailed Distribution](https://mlsanity.com/paper/2104.02703/) (2021)
- [Practical Adversarial Attacks on Spatiotemporal Traffic Forecasting Models](https://mlsanity.com/paper/2210.02447/) (2022)
- [Uncertainty Quantification via Spatial-Temporal Tweedie Model for Zero-inflated and Long-tail Travel Demand Prediction](https://mlsanity.com/paper/2306.09882/) (2023)
- [FAIR: Fair Adversarial Instance Re-weighting](https://mlsanity.com/paper/2011.07495/) (2020)
- [Revisiting Adversarial Training under Long-Tailed Distributions](https://mlsanity.com/paper/2403.10073/) (2024)
- [Enhancing the Robustness via Adversarial Learning and Joint Spatial-Temporal Embeddings in Traffic Forecasting](https://mlsanity.com/paper/2208.03063/) (2022)
- [It Is All About Data: A Survey on the Effects of Data on Adversarial Robustness](https://mlsanity.com/paper/2303.09767/) (2023)
- [Robust Spatiotemporal Traffic Forecasting with Reinforced Dynamic Adversarial Training](https://mlsanity.com/paper/2306.14126/) (2023)
- [Uncertainty-Aware Probabilistic Graph Neural Networks for Road-Level Traffic Accident Prediction](https://mlsanity.com/paper/2309.05072/) (2023)
- [Interpretable Cascading Mixture-of-Experts for Urban Traffic Congestion Prediction](https://mlsanity.com/paper/2406.12923/) (2024)
- [Improving Adversarial Robustness with Self-Paced Hard-Class Pair Reweighting](https://mlsanity.com/paper/2210.15068/) (2022)
- [Adversarial Diffusion Attacks on Graph-based Traffic Prediction Models](https://mlsanity.com/paper/2104.09369/) (2021)
- [SUSTeR: Sparse Unstructured Spatio Temporal Reconstruction on Traffic Prediction](https://mlsanity.com/paper/2502.16935/) (2025)

## Built on by (most cited first, 2 of 2)

- [Learning with Calibration: Exploring Test-Time Computing of Spatio-Temporal Forecasting](https://mlsanity.com/paper/2506.00635/) (2025)
- [Scaling Up AI-Generated Image Detection with Generator-Aware Prototypes](https://mlsanity.com/paper/2512.12982/) (2025)

## Abstract (arXiv)

Spatiotemporal Graph Learning (SGL) under Zero-Inflated Distribution (ZID) is crucial for urban risk management tasks, including crime prediction and traffic accident profiling. However, SGL models are vulnerable to adversarial attacks, compromising their practical utility. While adversarial training (AT) has been widely used to bolster model robustness, our study finds that traditional AT exacerbates performance disparities between majority and minority classes under ZID, potentially leading to irreparable losses due to underreporting critical risk events. In this paper, we first demonstrate the smaller top-k gradients and lower separability of minority class are key factors contributing to this disparity. To address these issues, we propose MinGRE, a framework for Minority Class Gradients and Representations Enhancement. MinGRE employs a multi-dimensional attention mechanism to reweight spatiotemporal gradients, minimizing the gradient distribution discrepancies across classes. Additionally, we introduce an uncertainty-guided contrastive loss to improve the inter-class separability and intra-class compactness of minority representations with higher uncertainty. Extensive experiments demonstrate that the MinGRE framework not only significantly reduces the performance disparity across classes but also achieves enhanced robustness compared to existing baselines. These findings underscore the potential of our method in fostering the development of more equitable and robust models.