1.1P Basic Linux Security
SIT719 Security and Privacy Issues in Analytics
Pass Task 7.1: Taxonomy of Attacks, Defenses, and
Consequences in Adversarial Machine Learning
Overview
The Information Technology Laboratory (ITL) at the National Institute of Standards and Technology
(NIST) promotes the U.S. economy and public welfare by providing technical leadership for the
Nation’s measurement and standards infrastructure. Recently NIST has published an internal
eport on “A Taxonomy and Terminology of Adversarial Machine Learning” (link below). This NIST
Interagency/Internal Report (NISTIR) is intended as a step toward securing applications of Artificial
Intelligence (AI), especially against adversarial manipulations of Machine Learning (ML), by
developing a taxonomy and terminology of Adversarial Machine Learning (AML).
Link: https:
nvlpubs.nist.gov/nistpubs/i
2019/NIST.IR.8269-draft.pdf
Please see the details of the task in the Task Description section.
This is a Pass task, so you MUST complete the task and submit the evidence of your work to
Ontrack.
Task Description
Suppose you are working in an organization who are developing a report on the vulnerabilities of
machine learning models due to adversarial attacks. Your manager has asked you to provide a
600 word report to submit within the next week. His expectation is that the 600 word report will
cover the attack taxonomies, defense mechanisms and consequences.
Instructions:
1. Read the NIST article from the below link:
https:
nvlpubs.nist.gov/nistpubs/i
2019/NIST.IR.8269-draft.pdf
2. Identify five important attack types. Summarize in approx. 300 words.
https:
nvlpubs.nist.gov/nistpubs/i
2019/NIST.IR.8269-draft.pdf
https:
nvlpubs.nist.gov/nistpubs/i
2019/NIST.IR.8269-draft.pdf
Hint: The above figure demonstrates the attack categories. It has been obtained from Figure
2 of the report.
2. Summarize the defense mechanisms for the attack types you identified in step 1. (Approx.
200 words)
Submit the report PDF to the OnTrack system.
Overview
Task Description