expertise needed
privacy engineering
Requires knowledge of privacy-enhancing methods and privacy threat modelling: disclosure and re-identification risk, privacy budgets, and the utility cost of protection.
14 techniques
| Goals | Models | Data Types | Description | |||
|---|---|---|---|---|---|---|
| GAN-Based Tabular Synthetic Data | Algorithmic | Architecture/neural Networks/generative/gan Architecture/neural Networks/generative/vae +3 | Tabular | GAN-based tabular synthetic data generation trains a generative model on a real table, then samples new rows from it to... | ||
| Sequential CART Synthesis | Algorithmic | Architecture/model Agnostic Requirements/black Box | Tabular | Sequential CART synthesis creates an artificial copy of a confidential microdata set (records about individual people or... | ||
| Federated Learning | Algorithmic | Architecture/linear Models Architecture/neural Networks +4 | Any | Federated learning trains one shared model across several parties, such as phones or hospitals, without collecting their... | ||
| Differential Privacy | Algorithmic | Architecture/model Agnostic Requirements/black Box | Any | Differential privacy puts a mathematical limit on what a published result reveals about any one person. The data holder... | ||
| Machine Unlearning | Algorithmic | Architecture/model Agnostic Architecture/neural Networks +2 | Any | Machine unlearning removes the influence of specific training records from a trained model, so that the model behaves as... | ||
| Membership Inference Attack Testing | Testing | Architecture/model Agnostic Requirements/black Box | Any | Membership inference attack testing checks whether someone with access to a trained model can tell if a particular... | ||
| Synthetic Data Evaluation | Testing | Architecture/model Agnostic Requirements/black Box | Tabular | Synthetic data evaluation checks whether an artificial dataset made to stand in for sensitive real records is safe to... | ||
| Canary Exposure Testing | Testing | Architecture/neural Networks Paradigm/generative +1 | Text | Canary exposure testing measures how strongly a training procedure makes a generative sequence model, such as a language... | ||
| Differential Privacy Auditing | Testing | Architecture/model Agnostic Requirements/white Box | Any | Differential privacy auditing tests whether a training procedure that claims a differential privacy guarantee at a... | ||
| Training Data Extraction Testing | Testing | Architecture/neural Networks/transformer/llm Paradigm/generative +1 | Text | Training data extraction testing measures how much of its training text a generative language model reproduces word for... | ||
| Attacker-Model Re-identification Risk Measurement | Metric | Architecture/model Agnostic Requirements/black Box | Tabular | Attacker-model re-identification risk measurement estimates, for a table of real records about people that is about to... | ||
| Differentially Private Stochastic Gradient Descent | Algorithmic | Architecture/neural Networks Paradigm/parametric +2 | Any | Differentially private stochastic gradient descent (DP-SGD) trains a neural network, or another model fitted by gradient... | ||
| k-Anonymity | Algorithmic | Architecture/model Agnostic Requirements/black Box | Tabular | k-Anonymity is a property that a table of real records about people can be made to satisfy before it is released or... | ||
| PII Detection and Redaction in Training Text | Algorithmic | Architecture/model Agnostic Architecture/neural Networks/transformer/llm +1 | Text | PII detection and redaction finds personally identifiable information (PII), such as names, email addresses, phone... |
Rows per page
Page 1 of 1