Dive into the essentials of ML model development, processes, and techniques to combat underfitting and overfitting, explore distributed training approaches, and understand model explainability. Enhance your skills with practical insights from a seasoned expert.

1. Session 3
Dr. Ryan Rad
Professional Machine Learning Engineer

2. Where are we on our journey
1
Session 3 Content Review
2
Extra Information
4
Next Steps
5
Sample Question Review
3
6 Q&A

3. Where are we on our
journey

4. Professional Machine Learning Certification
Learning Journey Organized by Google Developer Groups Surrey co hosting with GDG Seattle
Session 1
Feb 24, 2024
Virtual
Session 2
Mar 2, 2024
Virtual
Session 3
Mar 9, 2024
Virtual
Session 4
Mar 16, 2024
Virtual
Session 5
Mar 23, 2024
Virtual
Session 6
Apr 6, 2024
Virtual Review the
Professional ML
Engineer Exam
Guide
Review the
Professional ML
Engineer Sample
Questions
Go through:
Google Cloud
Platform Big Data
and Machine
Learning
Fundamentals
Hands On Lab
Practice:
Perform
Foundational Data,
ML, and AI Tasks in
Google Cloud
(Skill Badge) - 7hrs
Build and Deploy ML
Solutions on Vertex
AI
(Skill Badge) - 8hrs
Self
study
(and
potential
exam)
Lightning talk +
Kick-off & Machine
Learning Basics +
Q&A
Lightning talk +
GCP- Tensorflow &
Feature Engineering
+ Q&A
Lightning talk +
Enterprise Machine
Learning + Q&A
Lightning talk +
Production Machine
Learning with
Google Cloud + Q&A
Lightning talk + NLP
& Recommendation
Systems on GCP +
Q&A
Lightning talk + MOPs
& ML Pipelines on GCP
+ Q&A
Complete course:
Introduction to AI and
Machine Learning on
Google Cloud
Launching into
Machine Learning
Complete course:
TensorFlow on Google
Cloud
Feature
Engineering
Complete course:
Machine Learning in
the Enterprise
Hands On Lab
Practice:
Production Machine
Learning Systems
Computer Vision
Fundamentals with
Google Cloud
Complete course:
Natural Language
Processing on Google
Cloud
Recommendation
Systems on GCP
Complete course:
ML Ops - Getting
Started
ML Pipelines on Google
Cloud
Check Readiness:
Professional ML
Engineer Sample
Questions

5. Session 3 Content Review

6. Session 3
Study Group
Model Development
- Build a model.
- Train a model.
- Test a model.
- Scale model training and serving.
- Model Explainability

7. ML Processes

8. ML Processes

9. Underfit vs Overfit

10. Model Complexity vs Error

11. Model Complexity vs Error
Optimal Region

12. Model Complexity vs Error
Overfitting
Underfitting

13. Overfitting
How to fix/prevent overfitting:
● Add regularization
○ L1/L2 Reg
○ Dropout
● Reduce model complexity

14. Underfitting
How to fix/prevent underfitting:
● Lower regularization
coefficients
● Increase model complexity
● Feature Engineering

15. Data Catalog

18. Distributed Training Approaches
Async Parameter Server vs. Sync AllReduce

19. PS0 PS1
Worker 0 Worker 1 Worker 2 Worker 3
Async parameter server

20. PS0 PS1
Worker 0 Worker 1 Worker 2 Worker 3
x x x
Async parameter server

21. PS0 PS1
Worker 0 Worker 1 Worker 2 Worker 3
Async parameter server

22. Worker 0
Worker 3 Worker 2
Worker 1
p0, p1.. p0, p1..
p0, p1.. p0, p1..
Sync allreduce architecture

23. Consider async parameter server
if...
Consider sync allreduce if...
Many low-power or unreliable
workers.
Multiple devices on one host.
Fast devices with strong links (e.g.
TPUs).
Better for multiple GPUs.
Constrained by compute power.
More mature approach.
Constrained by I/O.
There isn’t one right answer, but here are some
considerations

24. Model Explainability

25. Gradient-based Attribution
attribution for
feature
Create attribution using gradient of the output wrt each base input
feature
● same as feature weights for linear models
● 1st order approximation for non-linear models
● use (normalized) attribution as mask/window over image

26. score
intensity
Interesting
gradients
uninterestin
g
gradients
1.0
0.0
Scaled
images
Scaled
gradients

27. Image Explanation

28. Sample Questions Review

29. Model complexity often refers to the number of features or terms included in a given predictive
model. What happens when the complexity of the model increases?
A. Model performance on a test set is going to be poor.
B. All of the options are correct.
C. Model is more likely to overfit.
D. Model will not figure out general relationships in the data.

30. Model complexity often refers to the number of features or terms included in a given predictive
model. What happens when the complexity of the model increases?
A. Model performance on a test set is going to be poor.
B. All of the options are correct.
C. Model is more likely to overfit.
D. Model will not figure out general relationships in the data.

31. You need to write a generic test to verify whether Dense Neural Network (DNN)
models automatically released by your team have a sufficient number of
parameters to learn the task for which they were built. What should you do?
A. Train the model for a few iterations, and check for NaN values.
B. Train the model for a few iterations, and verify that the loss is constant.
C. Train a simple linear model, and determine if the DNN model outperforms it.
D. Train the model with no regularization, and verify that the loss function is
close to zero.

32. You need to write a generic test to verify whether Dense Neural Network (DNN)
models automatically released by your team have a sufficient number of
parameters to learn the task for which they were built. What should you do?
A. Train the model for a few iterations, and check for NaN values.
B. Train the model for a few iterations, and verify that the loss is constant.
C. Train a simple linear model, and determine if the DNN model outperforms it.
D. Train the model with no regularization, and verify that the loss function is
close to zero.

33. You work for a textile manufacturer and have been asked to build a model to detect and classify
fabric defects. You trained a machine learning model with high recall based on high resolution
images taken at the end of the production line. You want quality control inspectors to gain trust in
your model. Which technique should you use to understand the rationale of your classifier?
A. Use K-fold cross validation to understand how the model performs on different test
datasets.
B. Use the Integrated Gradients method to efficiently compute feature attributions for each
predicted image.
C. Use PCA (Principal Component Analysis) to reduce the original feature set to a smaller set of
easily understood features.
D. Use k-means clustering to group similar images together, and calculate the Davies-Bouldin
index to evaluate the separation between clusters.

34. You work for a textile manufacturer and have been asked to build a model to detect and classify
fabric defects. You trained a machine learning model with high recall based on high resolution
images taken at the end of the production line. You want quality control inspectors to gain trust in
your model. Which technique should you use to understand the rationale of your classifier?
A. Use K-fold cross validation to understand how the model performs on different test
datasets.
B. Use the Integrated Gradients method to efficiently compute feature attributions for each
predicted image.
C. Use PCA (Principal Component Analysis) to reduce the original feature set to a smaller set of
easily understood features.
D. Use k-means clustering to group similar images together, and calculate the Davies-Bouldin
index to evaluate the separation between clusters.

35. Q&A

36. Preview actions for next
week

37. By our next meeting
1. Complete
a. Production Machine Learning Systems
b. Computer Vision Fundamentals with Google Cloud

38. Link to badge
Redeem your participation badge
Thank you for joining the event

39. Thank you for
tuning in!
For any operational questions about access to
Cloud Skills Boost or the Road to Google
Developers Certification program contact: gdg-
support@google.com