Safety
36 techniques
Ensuring AI systems operate safely and do not cause harm.
Browse by subcategory
All techniques
36 techniques
| Goals | Models | Data Types | Description | |||
|---|---|---|---|---|---|---|
| Out-of-Distribution Detector for Neural Networks | Algorithmic | Architecture/neural Networks Paradigm/discriminative +3 | Any | ODIN (Out-of-Distribution Detector for Neural Networks) identifies when a neural network encounters inputs significantly... | ||
| Simulation-Based Synthetic Data Generation | Algorithmic | Requirements/model Agnostic Paradigm/generative +1 | Tabular Time-series | Generates synthetic datasets through computational simulation of underlying data-generating processes, encompassing... | ||
| Federated Learning | Algorithmic | Architecture/linear Models Architecture/neural Networks +4 | Any | Federated learning enables collaborative model training across multiple distributed parties (devices, organisations, or... | ||
| Homomorphic Encryption | Algorithmic | Architecture/linear Models Architecture/model Agnostic +4 | Any | Homomorphic encryption allows computation on encrypted data without decrypting it first, producing encrypted results... | ||
| Deep Ensembles | Algorithmic | Architecture/neural Networks Paradigm/parametric +2 | Any | Deep ensembles combine predictions from multiple neural networks trained independently with different random... | ||
| Cross-validation | Algorithmic | Architecture/model Agnostic Paradigm/supervised +2 | Any | Cross-validation evaluates model performance and robustness by systematically partitioning data into multiple subsets... | ||
| Safety Envelope Testing | Testing | Architecture/model Agnostic Requirements/black Box | Any | Safety envelope testing systematically evaluates AI system performance at the boundaries of its intended operational... | ||
| Internal Review Boards | Process | Architecture/model Agnostic Requirements/black Box | Any | Internal Review Boards (IRBs) provide independent, systematic evaluation of AI/ML projects throughout their lifecycle to... | ||
| Red Teaming | Procedural | Architecture/model Agnostic Requirements/black Box | Any | Red teaming is a structured adversarial evaluation process in which a dedicated team systematically probes an AI/ML... | ||
| Human-in-the-Loop Safeguards | Process | Architecture/model Agnostic Requirements/black Box | Any | Human-in-the-loop safeguards establish systematic checkpoints where human experts review, validate, or override AI/ML... | ||
| Confidence Thresholding | Algorithmic | Architecture/model Agnostic Requirements/black Box +1 | Any | Confidence thresholding creates decision boundaries based on model uncertainty scores, routing predictions into... | ||
| Runtime Monitoring and Circuit Breakers | Algorithmic | Architecture/model Agnostic Requirements/black Box | Any | Runtime monitoring and circuit breakers establish continuous surveillance of AI/ML systems in production, tracking... | ||
| Model Cards | Documentation | Architecture/model Agnostic Requirements/black Box | Any | Model cards are standardised documentation frameworks that systematically document machine learning models through... | ||
| Datasheets for Datasets | Documentation | Architecture/model Agnostic Requirements/black Box | Any | Datasheets for datasets establish comprehensive documentation standards for datasets, systematically recording creation... | ||
| MLflow Experiment Tracking | Process | Architecture/model Agnostic Requirements/black Box | Any | MLflow is an open-source platform that tracks machine learning experiments by automatically logging parameters, metrics,... | ||
| Data Version Control | Process | Architecture/model Agnostic Requirements/black Box | Any | Data Version Control (DVC) is a Git-like version control system specifically designed for machine learning data, models,... | ||
| Model Distillation | Algorithmic | Architecture/neural Networks Paradigm/parametric +3 | Any | Model distillation transfers knowledge from a large, complex model (teacher) to a smaller, more efficient model... | ||
| Model Pruning | Algorithmic | Architecture/neural Networks Paradigm/parametric +4 | Any | Model pruning systematically removes less important weights, neurons, or entire layers from neural networks to create... | ||
| Neuron Activation Analysis | Algorithmic | Architecture/neural Networks Requirements/model Internals +1 | Text | Neuron activation analysis examines the firing patterns of individual neurons in neural networks by probing them with... | ||
| Prompt Sensitivity Analysis | Experimental | Architecture/neural Networks/transformer/llm Paradigm/generative +1 | Text | Prompt Sensitivity Analysis systematically evaluates how variations in input prompts affect large language model... |
Rows per page
Page 1 of 2