A Causal View on Robustness of Neural Networks
Explore this paper's citation graph
Summary
A deep causal manipulation augmented model (deep CAMA) is designed which explicitly models possible manipulations on certain causes leading to changes in the observed effect and achieves disentangled representation which separates the representation of manipulations from those of other latent causes.
- Type
- preprint
- Published
- 2020-05-03
- Cited by
- 102
- References
- 59
- Access
- Open access
- OpenAlex
- https://openalex.org/W2970990331
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:202765557
Keywords
Robustness (evolution), Deep neural networks, Computer science, Artificial intelligence, Artificial neural network
References
- Intriguing properties of neural networks
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- Auto-Encoding Variational Bayes
- Making Things Happen: A Theory of Causal Explanation
- Causation, Prediction, and Search
- The generic viewpoint assumption in a framework for visual perception
- A theory of causal learning in children: causal maps and Bayes nets.
- Local Causal and Markov Blanket Induction for Causal Discovery and Feature Selection for Classification Part I: Algorithms and Empirical Evaluation
- Domain Adaptation under Target and Conditional Shift
- Learning Structured Output Representation using Deep Conditional Generative Models
- Adversarial examples in the physical world
- Domain Adaptation with Conditional Transferable Components
- Towards Evaluating the Robustness of Neural Networks
- Deep Variational Information Bottleneck
- Detecting Adversarial Samples from Artifacts
- Practical Black-Box Attacks against Machine Learning
- Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methods
- Towards Deep Learning Models Resistant to Adversarial Attacks
- beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework
- Advances in Variational Inference
Cited by
- Decoder-free Robustness Disentanglement without (Additional) Supervision
- Learning Causal Semantic Representation for Out-of-Distribution Prediction
- Proactive Pseudo-Intervention: Causally Informed Contrastive Learning For Interpretable Vision Models
- Generalizing to Unseen Domains: A Survey on Domain Generalization
- Embracing the Disharmony in Medical Imaging: A Simple and Effective Framework for Domain Adaptation
- Graph Domain Adaptation: A Generative View
- Bayesian Causal Inference for Real World Interactive Systems
- Adversarial for Good? How the Adversarial ML Community's Values Impede Socially Beneficial Uses of Attacks
- Counterfactual Adversarial Learning with Representation Interpolation
- Enhancing Model Robustness and Fairness with Causality: A Regularization Approach
- Pulling Up by the Causal Bootstraps: Causal Data Augmentation for Pre-training Debiasing
- A Convolutional Autoencoder Topology for Classification in High-Dimensional Noisy Image Datasets
- CaRTS: Causality-driven Robot Tool Segmentation from Vision and Kinematics Data
- Causal GraphSAGE: A robust graph method for classification based on causal sampling
- Causal Representation Learning for Out-of-Distribution Recommendation
- Causal Discovery and Injection for Feed-Forward Neural Networks
- Certified Robustness Against Natural Language Attacks by Causal Intervention
- On the Generalization and Adaption Performance of Causal Models
- A mechanism informed neural network for predicting machining deformation of annular parts
- Training a Resilient Q-network against Observational Interference
Related papers
- Comprehensive Analysis of Hyperdimensional Computing Against Gradient Based Attacks
- Progressive Diversified Augmentation for General Robustness of DNNs: A Unified Approach
- Exploring Architectural Ingredients of Adversarially Robust Deep Neural Networks
- Exploring Architectural Ingredients of Adversarially Robust Deep Neural Networks
- Interpretable Mesomorphic Networks for Tabular Data
- Towards Adversarial Robustness of Deep Vision Algorithms
- Adversarial Robustness of Deep Learning: Theory, Algorithms, and Applications
- Deep learning: Evolution and expansion
- Multi-Condition Training on Deep Convolutional Neural Networks for Robust Plant Diseases Detection