1 of 23

Lecture 1:

Course Introduction

Soo Kyung Kim

Spring 2026

2 of 23

The Instructor

Spring 2026

2

3 of 23

The Instructor – Research Vision

Control

Science

  • Climate
  • Material

XAI

Foundational AI

Science & Engineering �Practices

�Sustainable Future

LLM�MLLM

RL

  • Reliability
  • Explainability
  • Generalizability

Spring 2026

3

4 of 23

The Instructor – Toward Sustainable Future

Foundational� AI

Science (Climate, Material)

Fasten Expensive Scientific Simulation

Discovering Scientific Pattern from Data

  • Save Energy
  • Fasten Decisions and Scientific Discovery

Control�

  • Efficient and Trustworthy Control
  • Save Energy by Optimizing System

Find human Interpretable� Symbolic Policy

Spring 2026

4

5 of 23

Office

  • Location: ECC 129-1
  • Students are always welcome to visit, for any topic!
    • Any question regarding the course, e.g., lectures, homework, ...
    • Research discussion
    • Career development
    • Chatting about relationships, life, and having fun together
    • … and more!
  • Appointment is recommended.
    • Email sookim@ewha.ac.kr to reserve your spot.
    • Walk-in is also fine, if I am in the office without meeting.

Spring 2026

5

6 of 23

Course Logistics

Spring 2026

6

7 of 23

Course Workflow

  • Course syllabus: https://eureka.ewha.ac.kr/eureka/my/hssg4008q.do?YEAR=2026&TERM_CD=10&GROUP_CD=20&SUBJECT_CD=39150&CLASS_NUM=01
  • Recommended reading list is available for each lecture.
    • Slides access will be given 1~2 days before the class.
    • Homeworks will be available on ETL.
  • Textbook
    • "Probabilistic Machine Learning: An Introduction (2nd Ed.)" by Kevin Murphy, 2021, MIT Press.
    • "Deep Learning" by Ian Goodfellow, Yoshua Bengio, Aaron Courville, 2015, MIT Press.
  • Course website
    • https://agi.ewha.ac.kr/teaching

Spring 2026

7

8 of 23

Topics to be covered (first half)

Spring 2026

8

9 of 23

Topics to be covered (second half)

Spring 2026

9

10 of 23

Lab Sessions

  • At each lab session, you will have a chance to practice implementing ML/DL models we cover in the class.�
  • lab sessions will be lecture-based.
    • No pre-recorded videos to watch.
    • No group activities.
    • We will basically walkthrough you to implement the models we learn in the class.
    • Lab sessions will be in-class.
    • You may bring your laptop to try programming on your own.

Spring 2026

10

11 of 23

Prerequisites

  • Fundamental Python programming skills
    • Should know the basic functionalities of Python.
    • Basic experience in numpy, Pytorch (or TensorFlow).�
  • Basic calculus, linear algebra, probability
    • Logarithm
    • Differentiation, chain rule
    • Matrix/tensor terminologies, concepts, operations
    • Gaussian distribution, mean and variance�
  • Data structures and algorithms
    • Time/space complexity, Big O notation

Spring 2026

11

12 of 23

Course Rules

  • No late submission accepted.
    • We accept homework submissions until 30 min after the published deadline without penalty.
    • After the 30 min of grace period, no late submission accepted unless we notify otherwise.
    • Exceptions are allowed only when arranged with the instructor at least 1 week before the deadline, with an understandable reason and evidence (e.g., conference attendance).�
  • Do NOT directly share your work and source code with anyone else.
    • Copying other students’ work or program code is a serious violation of student code of conduct.
    • High-level discussion is allowed and encouraged.�
  • Final grade is not negotiable for any reason.

Spring 2026

12

13 of 23

Grading

  • Components
    • �

​

​

​

​

​

  • Grading is based on relative score.

Spring 2026

13

14 of 23

Introduction to AGI

Spring 2026

14

15 of 23

Why Data Science & Machine Learning & AGI?

Spring 2026

15

16 of 23

Abstraction & Knowledge

Archimedes (BC 287~212)

Newton (1643~1727)

Mostly relying on human intuition

Spring 2026

16

17 of 23

Abstraction & Knowledge

Out mind’s capacity, considered constant over time.

Spring 2026

17

18 of 23

Abstraction & Knowledge

Statistics (19~20C)

Big data (21C)

Large-scale data far beyond human intuition, requiring Machine Learning & high performance computing!

More systematic approach with mathematical tools

Spring 2026

18

19 of 23

History of AI: Before Deep Learning Revolution

Neural network ideas proposed

Backpropagation

Big data + GPU

ENIAC (1946)

Alan Turing, Computing Machinery and Intelligence (1950)

Spring 2026

19

20 of 23

Deep Learning Revolution

Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton. ImageNet Classification with Deep Convolutional Neural Networks (AlexNet), NIPS 2012. [paper]

AD 1

2

3

4

5

6

BC 1

BC 2

Spring 2026

20

21 of 23

Deep Learning Revolution

Geoffrey E. Hinton (Turing Award 2018)

야밤의 공대생 만화, https://goo.gl/mbD4cl

Spring 2026

21

22 of 23

Deep Learning Revolution

https://arxiv.org/pdf/1905.05055.pdf

Spring 2026

22

23 of 23

This Course

Machine Learning & Data Mining

  • Foundation of machine learning principles and methodologies in the first half
  • Covers basics of advanced data mining approaches in the second half
  • Equipped with lab sessions for actual implementation.

Spring 2026

23