Identify Susceptible Locations in Medical Records via Adversarial Attacks on Deep Predictive Models
Explore this paper's citation graph
Summary
This paper proposes an efficient and effective framework that learns a time-preferential minimum attack targeting the LSTM model with EHR inputs, and leverages this attack strategy to screen medical records of patients and identify susceptible events and measurements.
- Type
- article
- Published
- 2018-02-13
- Cited by
- 67
- References
- 43
- Access
- Open access
- OpenAlex
- https://openalex.org/W2787487383
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:3274404
Keywords
Computer science, Leverage (statistics), Adversarial system, Deep learning, Context (archaeology)
References
- Intriguing properties of neural networks
- Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
- From micro to macro: data driven phenotyping by densification of longitudinal electronic medical records
- Deep Computational Phenotyping
- Long Short-Term Memory
- Towards heterogeneous temporal clinical event pattern discovery: a convolutional approach
- Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
- Doctor AI: Predicting Clinical Events via Recurrent Neural Networks
- The Limitations of Deep Learning in Adversarial Settings
- DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks
- Multi-layer Representation Learning for Medical Concepts
- Crafting adversarial input sequences for recurrent neural networks
- Adversarial Diversity and Hard Positive Generation
- MIMIC-III, a freely accessible critical care database
- Deep Patient: An Unsupervised Representation to Predict the Future of Patients from the Electronic Health Records
- Deepr: A Convolutional Net for Medical Records
- Towards Evaluating the Robustness of Neural Networks
- DeepDGA: Adversarially-Tuned Domain Generation and Detection
- Universal Adversarial Perturbations
- Understanding Neural Networks through Representation Erasure
Cited by
- Defend Deep Neural Networks Against Adversarial Examples via Fixed andDynamic Quantized Activation Functions
- Parametric Noise Injection: Trainable Randomness to Improve Deep Neural Network Robustness Against Adversarial Attack
- Adversarial Examples for Hamming Space Search
- Generating Textual Adversarial Examples for Deep Learning Models: A Survey
- Longitudinal Adversarial Attack on Electronic Health Records Data
- Exploiting the Vulnerability of Deep Learning-Based Artificial Intelligence Models in Medical Imaging: Adversarial Attacks
- POBA-GA: Perturbation Optimized Black-Box Adversarial Attacks via Genetic Algorithm
- MetaPred: Meta-Learning for Clinical Risk Prediction with Limited Patient Electronic Health Records
- Is Robustness the Cost of Accuracy? - A Comprehensive Study on the Robustness of 18 Deep Image Classification Models
- Towards Robust and Discriminative Sequential Data Learning: When and How to Perform Adversarial Training?
- Query-Efficient Black-Box Attack by Active Learning
- AI in Health: State of the Art, Challenges, and Future Directions
- Cross-Modal Learning with Adversarial Samples
- Universal Adversarial Perturbation for Text Classification
- AdaCare: Explainable Clinical Health Status Representation Learning via Scale-Adaptive Feature Extraction and Recalibration
- Adversarial Attacks on Deep-learning Models in Natural Language Processing
- Recent Advances on Graph Analytics and Its Applications in Healthcare
- Vulnerability vs. Reliability: Disentangled Adversarial Examples for Cross-Modal Learning
- Adversarial Attacks to Machine Learning-Based Smart Healthcare Systems
- Generalizing Universal Adversarial Attacks Beyond Additive Perturbations
Related papers
- Global Adversarial Attacks for Assessing Deep Learning Robustness
- Developing and Defeating Adversarial Examples
- Generating adversarial examples for DNN using pooling layers
- Adversarial Perturbation Defense on Deep Neural Networks
- A Brief Comparison Between White Box, Targeted Adversarial Attacks in Deep Neural Networks
- Efficient Defenses Against Adversarial Attacks