harvard-edge · mpstewart1 · Nov 30, 2023 · Nov 21, 2023 · Nov 22, 2023 · Nov 22, 2023
diff --git a/images/cover_security_privacy.png b/images/cover_security_privacy.png
diff --git a/images/security_privacy/image1.png b/images/security_privacy/image1.png
diff --git a/images/security_privacy/image10.png b/images/security_privacy/image10.png
diff --git a/images/security_privacy/image11.png b/images/security_privacy/image11.png
diff --git a/images/security_privacy/image12.png b/images/security_privacy/image12.png
diff --git a/images/security_privacy/image13.png b/images/security_privacy/image13.png
diff --git a/images/security_privacy/image14.png b/images/security_privacy/image14.png
diff --git a/images/security_privacy/image15.png b/images/security_privacy/image15.png
diff --git a/images/security_privacy/image16.png b/images/security_privacy/image16.png
diff --git a/images/security_privacy/image2.png b/images/security_privacy/image2.png
diff --git a/images/security_privacy/image3.png b/images/security_privacy/image3.png
diff --git a/images/security_privacy/image4.png b/images/security_privacy/image4.png
diff --git a/images/security_privacy/image5.png b/images/security_privacy/image5.png
diff --git a/images/security_privacy/image6.png b/images/security_privacy/image6.png
diff --git a/images/security_privacy/image7.png b/images/security_privacy/image7.png
diff --git a/images/security_privacy/image8.png b/images/security_privacy/image8.png
diff --git a/images/security_privacy/image9.png b/images/security_privacy/image9.png
diff --git a/ops.qmd b/ops.qmd
@@ -234,7 +234,7 @@ Tight coupling between ML model components makes isolating changes difficult. Mo
 
 ### Correction Cascades
 
-![Figure 14.3: The flowchart depicts the concept of correction cascades in the ML workflow, from problem statement to model deployment. The arcs represent the potential iterative corrections needed at each stage of the workflow, with different colors corresponding to distinct issues such as interacting with physical world brittleness, inadequate application-domain expertise, conflicting reward systems, and poor cross-organizational documentation. The red arrows indicate the impact of cascades, which can lead to significant revisions in the model development process, while the dotted red line represents the drastic measure of abandoning the process to restart. This visual emphasizes the complex, interconnected nature of ML system development and the importance of addressing these issues early in the development cycle to mitigate their amplifying effects downstream. [@data_cascades]](images/ai_ops/data_cascades.png)
+![Figure 14.3: The flowchart depicts the concept of correction cascades in the ML workflow, from problem statement to model deployment. The arcs represent the potential iterative corrections needed at each stage of the workflow, with different colors corresponding to distinct issues such as interacting with physical world brittleness, inadequate application-domain expertise, conflicting reward systems, and poor cross-organizational documentation. The red arrows indicate the impact of cascades, which can lead to significant revisions in the model development process, while the dotted red line represents the drastic measure of abandoning the process to restart. This visual emphasizes the complex, interconnected nature of ML system development and the importance of addressing these issues early in the development cycle to mitigate their amplifying effects downstream. [@sculley2015hidden](images/ai_ops/data_cascades.png)
 
 Building models sequentially creates risky dependencies where later models rely on earlier ones. For example, taking an existing model and fine-tuning it for a new use case seems efficient. However, this bakes in assumptions from the original model that may eventually need correction. 
 

diff --git a/privacy_security.qmd b/privacy_security.qmd
diff --git a/references.bib b/references.bib