Publication Library
Adversarial Machine Learning A Taxonomy and Terminology of Attacks and Mitigations
Description: This NIST Trustworthy and Responsible AI report develops a taxonomy of concepts and defnes terminology in the feld of adversarial machine learning (AML). The taxonomy is built on surveying the AML literature and is arranged in a conceptual hierarchy that includes key types of ML methods and lifecycle stages of attack, attacker goals and objectives, and attacker capabilities and knowledge of the learning process. The report also provides corresponding methods for mitigating and managing the consequences of attacks and points out relevant open challenges to take into account in the lifecycle of AI systems. The terminology used in the report is consistent with the literature on AML and is complemented by a glossary that defnes key terms associated with the security of AI systems and is intended to assist non-expert readers. Taken together, the taxonomy and terminology are meant to inform other standards and future practice guides for assessing and managing the security of AI systems, by establishing a common language and understanding of the rapidly developing AML landscape.
Created At: 14 December 2024
Updated At: 14 December 2024
Data Classification Concepts and Considerations for Improving Data Collection
Description: Data classification is the process an organization uses to characterize its data assets using persistent labels so those assets can be managed properly. Data classification is vital for protecting an organization’s data at scale because it enables the application of cybersecurity and privacy protection requirements to the organization’s data assets. This publication defines basic terminology and explains fundamental concepts in data classification so there is a common language for all to use. It can also help organizations improve the quality and efficiency of their data protection approaches by becoming more aware of data classification considerations and taking them into account in business and mission use cases, such as secure data sharing, compliance reporting and monitoring, zero-trust architecture, and large language models.
Created At: 14 December 2024
Updated At: 14 December 2024
Secure Software Development Practices for Generative AI and Dual-Use Foundation Models
Description: This document augments the secure software development practices and tasks defined in Secure Software Development Framework (SSDF) version 1.1 by adding practices, tasks, recommendations, considerations, notes, and informative references that are specific to AI model development throughout the software development life cycle. These additions are documented in the form of an SSDF Community Profile to support Executive Order (EO) 14110, Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence, which tasked NIST with “developing a companion resource to the [SSDF] to incorporate secure development practices for generative AI and for dual-use foundation models.” This Community Profile is intended to be useful to the producers of AI models, the producers of AI systems that use those models, and the acquirers of those AI systems. This Profile should be used in conjunction with NIST Special Publication (SP) 800-218, Secure Software Development Framework (SSDF) Version 1.1: Recommendations for Mitigating the Risk of Software Vulnerabilities.
Created At: 14 December 2024
Updated At: 14 December 2024
Random Forests
Description: Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest. The generalization error for forests converges a.s. to a limit as the number of trees in the forest becomes large. The generalization error of a forest of tree classifiers depends on the strength of the individual trees in the forest and the correlation between them. Using a random selection of features to split each node yields error rates that compare favorably to Adaboost (Freund and Schapire[1996]), but are more robust with respect to noise. Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the splitting. Internal estimates are also used to measure variable importance. These ideas are also applicable to regression.
Created At: 14 December 2024
Updated At: 14 December 2024
Foundations of the Theory of Probability - Kolmogorov
Description: Foundations of the Theory of Probability - Kolmogorov
Created At: 14 December 2024
Updated At: 14 December 2024