Machine Learning Specialization · Unsupervised Learning, Recommenders, Reinforcement Learning

Anomaly Detection

A Gaussian anomaly detection system for server monitoring, using throughput and latency to identify abnormal machines by probability thresholding.

Gaussian density estimationMultivariate probabilityThreshold selectionF1 score validationOutlier detection

GitHub

What this project is about

The page is about turning probability into monitoring logic. Instead of labeling every server by hand, the project estimates what normal behavior looks like from throughput and latency, then flags low-probability machines as anomalies. The validation set is used to choose the threshold, so the system is not only visually intuitive but also measured by F1 score.

Implementation focus

I implemented Gaussian parameter estimation, probability scoring, and threshold search. The project also scales from the two-feature visualization to a higher-dimensional dataset, which makes it a useful bridge between a teaching example and a real monitoring-style workflow.

Key results and artifacts

Best validation F1: 0.875
Selected epsilon: 8.990853e-05
High-dimensional version flagged 117 anomalies

What I learned

This milestone connected probability theory to operational monitoring: estimate normal behavior first, then detect the rare cases that fall outside it.

Milestone deep dive

What this project proves in my learning path.

A Gaussian anomaly detection system for server monitoring, using throughput and latency to identify abnormal machines by probability thresholding. The milestone is included because it shows a complete learning unit: a problem, a data representation, an implementation path, measurable outputs, and a lesson that later connects to larger AI systems.

Input layerimages, anchors, boxes, and class scores

Model layerYOLO-style detection pipeline with filtering and NMS

Evaluation layertests, metrics, plots, or qualitative output

Artifact layerlocalized objects with confidence scores

Problem framing

What I was trying to solve

This page is not just a lab description. It explains how I transformed the assignment idea into an engineering story: what the model receives, what it optimizes, what output it should produce, and how the result can be inspected.

Implementation

What I coded or assembled

The implementation focus was the mechanism behind the result: functions, model blocks, training logic, preprocessing, evaluation, and the small technical details that make the notebook work rather than only look correct.

Concepts

Core concepts

Gaussian density estimationMultivariate probabilityThreshold selectionF1 score validationOutlier detection

These tags show the technical ideas this milestone added to my toolkit.

Learning result

What changed after it

This milestone connected probability theory to operational monitoring: estimate normal behavior first, then detect the rare cases that fall outside it.

Important outputs and evidence

Best validation F1: 0.875
Selected epsilon: 8.990853e-05
High-dimensional version flagged 117 anomalies

How I would explain it in an interview

I would describe this milestone by starting from the data representation, then explain the model or algorithm, then show what was measured. The important part is not only the final number; it is the ability to reason about why the method works, what assumptions it makes, and what limitations should be improved next.