Lead the GenAI revolution. Future-proof your skills by understanding AI basics, machine learning, algorithms, ethical practices, and real-world applications. Stay ahead in tech leadership.

In one form or the other, AI is going to be infused in all the tech products. Meaning, that in order to stay a relevant leader, it has become essential to have a solid, broad understanding of AI. While you may not be implementing the solution, you need to speak the language of AI.
 
Whether you’re leading a team on creating AI solutions, or you just want a better understanding of AI, then this course is for you. In this course, I have deconstructed the concepts most relevant for you.
 
You will start by learning the basics of AI, its connection with machine learning, and the different types of algorithms you should be familiar with. You’ll then learn the rules, best practices, and infrastructure for creating great AI products that users trust. Then, you will go through some real case studies of AI application and do a deep dive into Ethical AI. You'll end by getting an outlook into the current status and future of AI. In a nutshell, this course is your no-fuss, comprehensive guide to the essentials of AI.

Grokking AI for Engineering & Product Managers

# 3. AUC-ROC curve

AUC (area under the curve) - ROC (receiver operating characteristics) curve is a performance measurement for a classification model at various classification threshold settings. Basically, it is a probability curve that tells us how well the model is capable of distinguishing between classes. The higher the AUC value of our probability curve, the better the model is at predicting zeros as zeros and ones as ones.

**What do we mean by various threshold settings?**<br>
Say we set the threshold to 0.9. This means that if for any given sample our trained model predicts a value higher than 0.9, our output class will be predicted as a positive class. Otherwise, it will be placed in the negative class.

The ROC curve is plotted with the **true positive rate (TPR)** (Recall/Sensitivity) against the **false positive rate (FPR)** (FPR, *1 - Specificity*), where TPR is on the y-axis and FPR is on the x-axis, and:

- **Sensitivity, Recall, Hit Rate, or True Positive Rate**

  <img src="https://latex.codecogs.com/svg.latex?\Large&space;\mathrm{Sensitivity = TPR} = {\frac {\mathrm {TP} }{\mathrm {P} }}={\frac {\mathrm {TP} }{\mathrm {TP} +\mathrm {FN} }}"/> 

- **Fall-out (1 - Specificity), or False Positive Rate**

  <img src="https://latex.codecogs.com/svg.latex?\Large&space;\mathrm {1 - Specificity = FPR} ={\frac {\mathrm {FP} }{\mathrm {N} }}={\frac {\mathrm {FP} }{\mathrm {FP} +\mathrm {TN} }}"/> 

A great model has an AUC close to one, indicating it has an excellent measure of separability. On the other hand, a poor model has an AUC near zero, meaning it is predicting zeros as ones and ones as zeros. When AUC is 0.5, it means the model has no class separation capacity whatsoever, and it’s essentially making random predictions.

Let's understand this better via an example analysis taken from a medical research journal:



This lesson discusses the AUC-ROC curve and provides an example.


The Fundamentals

AI in Practice

Real Case Studies

Responsible AI

What's Next

Evaluation Metrics II

3. AUC-ROC curve