# A Causal View on Robustness of Neural Networks

arXiv: https://arxiv.org/abs/2005.01095 · HTML: https://mlsanity.com/paper/2005.01095/

## Key facts

- Cited by 53 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 May 2020.

## Built on by (most cited first, 30 of 53)

- [Generalizing to Unseen Domains: A Survey on Domain Generalization](https://mlsanity.com/paper/2103.03097/) (2021)
- [Backdoor Defense via Deconfounded Representation Learning](https://mlsanity.com/paper/2303.06818/) (2023)
- [CausalAdv: Adversarial Robustness through the Lens of Causality](https://mlsanity.com/paper/2106.06196/) (2021)
- [Graph Domain Adaptation: A Generative View](https://mlsanity.com/paper/2106.07482/) (2021)
- [The Essential Role of Causality in Foundation World Models for Embodied AI](https://mlsanity.com/paper/2402.06665/) (2024)
- [Out-of-distribution Prediction with Invariant Risk Minimization: The Limitation and An Effective Fix](https://mlsanity.com/paper/2101.07732/) (2021)
- [Certified Robustness Against Natural Language Attacks by Causal Intervention](https://mlsanity.com/paper/2205.12331/) (2022)
- [Explicit Tradeoffs between Adversarial and Natural Distributional Robustness](https://mlsanity.com/paper/2209.07592/) (2022)
- [Learning Attention Propagation for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2210.11557/) (2022)
- [Embracing the Disharmony in Medical Imaging: A Simple and Effective Framework for Domain Adaptation](https://mlsanity.com/paper/2103.12857/) (2021)
- [VACA: Design of Variational Graph Autoencoders for Interventional and Counterfactual Queries](https://mlsanity.com/paper/2110.14690/) (2021)
- [Domain-Specific Bias Filtering for Single Labeled Domain Generalization](https://mlsanity.com/paper/2110.00726/) (2021)
- [Interpreting the Robustness of Neural NLP Models to Textual Perturbations](https://mlsanity.com/paper/2110.07159/) (2021)
- [RS2G: Data-Driven Scene-Graph Extraction and Embedding for Robust Autonomous Perception and Scenario Understanding](https://mlsanity.com/paper/2304.08600/) (2023)
- [CaRTS: Causality-driven Robot Tool Segmentation from Vision and Kinematics Data](https://mlsanity.com/paper/2203.09475/) (2022)
- [Adversarial Robustness for Tabular Data through Cost and Utility Awareness](https://mlsanity.com/paper/2208.13058/) (2022)
- [A Review of the Role of Causality in Developing Trustworthy AI Systems](https://mlsanity.com/paper/2302.06975/) (2023)
- [Adversarial Visual Robustness by Causal Intervention](https://mlsanity.com/paper/2106.09534/) (2021)
- [Enhancing Model Robustness and Fairness with Causality: A Regularization Approach](https://mlsanity.com/paper/2110.00911/) (2021)
- [Deep Causal Learning: Representation, Discovery and Inference](https://mlsanity.com/paper/2211.03374/) (2022)
- [Emerging Synergies in Causality and Deep Generative Models: A Survey](https://mlsanity.com/paper/2301.12351/) (2023)
- [Enhancing Adversarial Contrastive Learning via Adversarial Invariant Regularization](https://mlsanity.com/paper/2305.00374/) (2023)
- [Training a Resilient Q-Network against Observational Interference](https://mlsanity.com/paper/2102.09677/) (2021)
- [A.I. Robustness: a Human-Centered Perspective on Technological Challenges and Opportunities](https://mlsanity.com/paper/2210.08906/) (2022)
- [Disentanglement and Generalization Under Correlation Shifts](https://mlsanity.com/paper/2112.14754/) (2021)
- [On the Generalization and Adaption Performance of Causal Models](https://mlsanity.com/paper/2206.04620/) (2022)
- [Moderately Distributional Exploration for Domain Generalization](https://mlsanity.com/paper/2304.13976/) (2023)
- [Instrumental Variable-Driven Domain Generalization with Unobserved Confounders](https://mlsanity.com/paper/2110.01438/) (2021)
- [Pulling Up by the Causal Bootstraps: Causal Data Augmentation for Pre-training Debiasing](https://mlsanity.com/paper/2108.12510/) (2021)
- [Causal Discovery and Knowledge Injection for Contestable Neural Networks (with Appendices)](https://mlsanity.com/paper/2205.09787/) (2022)

## Abstract (arXiv)

We present a causal view on the robustness of neural networks against input manipulations, which applies not only to traditional classification tasks but also to general measurement data. Based on this view, we design a deep causal manipulation augmented model (deep CAMA) which explicitly models possible manipulations on certain causes leading to changes in the observed effect. We further develop data augmentation and test-time fine-tuning methods to improve deep CAMA's robustness. When compared with discriminative deep neural networks, our proposed model shows superior robustness against unseen manipulations. As a by-product, our model achieves disentangled representation which separates the representation of manipulations from those of other latent causes.