JavaScript isn't enabled in your browser, so this file can't be opened. Enable and reload.

1 of 21

Welcome to Day 2 of

Introduction to Machine Learning Workshops

We’ll start at 1:00 PM as planned

2 of 21

Plan for the day …

Day 2
Activity	Start	End
Zoom room opens	12:45 PM	1:00 PM
What did we learn on Tuesday?	1:00 PM	1:20 PM
Lecture	1:20 PM	1:55 PM
First break	1:55 PM	2:10 PM
Lecture	2:10 PM	2:55 PM
Second break	2:55 PM	3:10 PM
Lecture	3:10 PM	3:55 PM
Conclusion	3:55 PM	4:00 PM

3 of 21

What did we learn on Tuesday?

A quick review

An Introduction

M. Ali Yousuf

maliyou1@umbc.edu

June 17, 18, 2025

For course material, see https://sites.google.com/view/aiandmlsummer2025/home

4 of 21

What is (Human) Intelligence?

https://www.merriam-webster.com/dictionary/intelligence

5 of 21

Run/Execute

Read/Edit

6 of 21

https://colab.research.google.com/drive/1bcKMWbpsBepG-96wbjMRahWyIF6-Y7Ha?usp=sharing

7 of 21

Three types of learning …

https://images.app.goo.gl/yaUmFoAmHcfeNA798

Supervised Learning

Unsupervised Learning

https://images.app.goo.gl/PkESdNzEMQxznFcg9

Reinforcement Learning

https://images.app.goo.gl/s9EBB19XLMcwL4t36

8 of 21

Three (common) types of learning …

Supervised	Regression	x1	x2	x3	x4	Y
		12.42	7.03	13.15	1.65	13.17
		9.65	7.80	13.90	1.14	7.39
		3.89	11.81	4.01	4.52	10.40
Supervised	Classification	10.37	1.48	12.10	5.88	A
		3.96	6.18	11.39	14.97	B
		5.01	2.35	10.67	4.90	B
		2.59	0.05	5.13	1.38	A
Unsupervised	Clustering	12.11	5.03	7.84	13.56
		4.96	13.61	1.40	7.95
		4.81	14.46	4.92	11.18
		9.20	7.98	4.71	0.20

9 of 21

Simple Regression

https://colab.research.google.com/drive/1FAp1Dhf1Yf9NueOfGCSgEXc3SP6kWJP7?usp=sharing

from sklearn.linear_model import LinearRegression

reg = LinearRegression()

reg.fit(df[['x']], df['y'])

y_pred = reg.predict(df[['x']])

Scroll down on the above page and now you will see the mathematics of Regression.
Notice how Cartesian Geometry, Calculus, Statistics and Programming blend together!

10 of 21

Classification (We used Logistic Regression which is, ironically, NOT Regression but Classification!)

Iris Flower (three types or classes)

11 of 21

from sklearn.linear_model import LogisticRegression

from sklearn.metrics import accuracy_score

model = LogisticRegression()

model.fit(X_train, y_train)

y_pred = model.predict(X_test)

accuracy = accuracy_score(y_test, y_pred)

Scroll down on the above page and now you will see the mathematics of Logistic Regression.
Notice how Statistics, Probability and Programming blend together!

https://colab.research.google.com/drive/1zAInwFDhcDc_kPnl-UVe5ePD6lZMvnUy?usp=sharing

12 of 21

Classification of Digits

For example, 0 is

[ 0. 0. 12. 9. 12. 1. 0. 0. 0. 0. 14. 16. 16. 8. 0. 0. 0. 3. 16. 9. 3. 15. 2. 0. 0. 4. 16. 1. 0. 16. 5. 0. 0. 5. 12. 0. 0. 16. 5. 0. 0. 3. 14. 1. 4. 16. 4. 0. 0. 0. 15. 12. 14. 14. 0. 0. 0. 0. 7. 12. 12. 2. 0. 0.]

See the full data here https://docs.google.com/spreadsheets/d/1hOjAbJ3TkoU3pHksq60_xfXwJ0neNyUD183Vp1Km-Eg/edit?usp=sharing

13 of 21

Sample #	X1	X2	X3	…	…	X63	X64	y
1	95	69	36	…	…	96	59	7
2	81	69	64	…	…	44	55	3
3	97	11	2	…	…	56	27	9
4	27	4	70	…	…	70	84	4
5	12	13	31	…	…	53	16	4
6	61	39	84	…	…	95	51	9
7	53	59	52	…	…	81	47	3
8	62	68	8	…	…	87	63	4


	72	98	52	…	…	73	8	6
	39	78	72	…	…	82	90	7
	39	2	18	…	…	47	95	3
	67	82	89	…	…	78	14	1
	46	55	34	…	…	53	1	1
1797	96	62	65	…	…	57	92	8

14 of 21

https://colab.research.google.com/drive/1hZZuPJgtX-1U1J_vOUN19iRpnX3PvdQh?usp=sharing

from sklearn.linear_model import Perceptron

Model = Perceptron(tol=1e-3, random_state=60)

Model.fit(X,y)

Scroll down on the above page and now you will see the mathematics of Classification.
Notice how Statistics, Probability and Programming blend together!

15 of 21

And what did we miss?

16 of 21

MPS in Data Science at UMBC Shady Grove

UMBC’s Master of Professional Studies (MPS) in Data Science program prepares students from a wide range of disciplinary backgrounds for careers in data science.
In 7 core courses, students will get a fundamental understanding of data science through classes that highlight machine learning, data analysis and data management.
Beyond the core courses, students will take 3 courses in domain specific pathways developed in collaboration with academic departments across the university and beyond.

https://professionalprograms.umbc.edu/data-science/masters-of-professional-studies-data-science/

Contact: M. A. Yousuf, maliyou1@umbc.edu

17 of 21

Academic Programs: MPS

DATA 601 Introduction to Data Science
DATA 602 Introduction to Data Analysis and Machine Learning
DATA 603 Platforms for Big Data Processing
DATA 604 Data Management
DATA 605 Ethical and Legal Issues in Data Science
DATA 606 Capstone in Data Science
ENMG 652 Management Leadership and Communications
Pathway Elective I
Pathway Elective II
Pathway Elective III

��Master of Professional Studies:

Data Science

10 courses

30 credits

18 of 21

Who am I?

I have extensive consulting experience in medical devices and AI in healthcare, and teaching experience in parallel computing, AI, robotics and intelligent manufacturing.

I hold

a Ph.D. in Theoretical High Energy Physics (Big Data) from Quaid-i-Azam University,
a postdoc in Parallel Supercomputing (Big Data) from SUNY Albany and
a postdoc in Medical Devices and Data from the University of Maryland, Baltimore.

Before coming to UMBC as the Graduate Program Director at Shady Grove:

Tec de Monterrey (Mexico City)
St. Agnes Cancer Institute
Johns Hopkins University
TurningPoint Healthcare

Find me on Twitter and LinkedIn

19 of 21

https://sites.google.com/view/pagita/home

20 of 21

https://sites.google.com/view/mlandaiinhealthcare/

21 of 21

https://quotefancy.com/quote/37364/Ellen-Johnson-Sirleaf-If-your-dreams-do-not-scare-you-they-are-not-big-enough