Course Logistics

(1)

Course Logistics

Applied Deep Learning

February 14nd, 2022 http://adl.miulab.tw

(2)

Course Logistics

◉

Instructor: 陳縕儂 Yun-Nung (Vivian) Chen

◉

Time: Monday 234, 9:10-12:20

◉

Location: 資103

◉

Website: http://adl.miulab.tw

◉

NTU COOL: https://cool.ntu.edu.tw/courses/14072

◉

Email: [email protected]

○

To ensure timely response, email title should contain “[ADL2022]”

○

Do NOT send to our personal emails

2

Always check the up-to-date information from the course website

(3)

NTU COOL for Fighting Coronavirus

◉

^{NTU COOL}

○

Lecture videos

■

Comments anytime

○

Assignment submission (還可以寫 code 呢!)

◉

^{Slido QA}

○

^#ADL2022

◉

TA Team

○

Forum discussion (preferred)

○

Email QA

○

TA recitation/hours (maybe virtual)

3

(4)

Course Goal

◉

The students are expected to understand

1.

how deep learning works

2.

how to frame tasks into learning problems

3.

how to use toolkits to implement designed models, and

4.

when and why specific deep learning techniques work for specific problems

4

(5)

Pre-requisites

◉

^Course

○

Required: college-level calculus, linear algebra

○

Preferred: probability, statistics

◉

Programming

○

proficiency in Python; all assignments will be in Python

○

GitHub; all assignments will be handed in via GitHub

○

Kaggle; all assignments will be submitted to Kaggle

5

(tutorialfrom Stanford) (tutorial) (website)

Please consider your available resources for taking this course

(6)

GitHub Student Pack

◉

The student plan provides unlimited private repositories

○

make your assignments private before the due date

○

make them public afterwards

6

(7)

Grading Policy

◉

3~4 Individual Assignment: 60%

○

GitHub code w/ README

■

The score is based on coding and the report

■

Bonus points for outstanding performance

■

Late policy: 25% off per day late afterwards

◉

Final Group Project: 35%

○

GitHub code, Project document

■

Bonus points for the outstanding work

○

Final presentation (format TBA)

◉

Participation: 5%

○

Forum/slido discussion involvement

○

Write-up for the special events

7

Understanding the difference between “collaboration” and “academic infraction”

(8)

Individual Assignments

8

A1. Sequence Tagging A2. Transformer / BERT A3. Language Generation

(9)

Final Group Project (2~5 persons)

◉

The final project topic will be announced later

○

Presentation

■

Poster or oral presentation

○

Peer grading

○

Project Report & Code

■

Wrap-up project report

■

GitHub code submission w/ README 9

The project details will be announced later

(10)

Tentative Schedule

10

Week Topic Assignment

1 2022/02/14 Course Logistics, Introduction, NN Basics A0 – Pytorch Tutorial 2 2022/02/21 Backpropagation, Word Representations, RNN A1 – Sequence Tagging

2022/02/28 Break

3 2022/03/07 Word Embeddings, Gating, Attention 4 2022/03/14 Transformer

5 2022/03/21 ELMo, BERT A2 – BERT

6 2022/03/28 More BERT 7 2022/04/04 Break

彈性補充 Deep RL, Value-Based RL A2 – BERT

8 2022/04/11 Midterm Break

彈性補充 Policy Gradient + Actor-Critic

9 2022/04/18 Natural Language Generation A3 – NLG 10 2022/04/25 Beyond Supervised Learning

11 2022/05/02 Towards Conversational AI

12 2022/05/09 E2E Conversational AI Final Project 13 2022/05/16 TBA

14 2022/05/23 TBA

15 2022/05/30 Final Project Presentation

(11)

Teaching Assistant Team

11

(12)

Rules

12

Asking questions is encouraged!!

Any comment or feedback is preferred!!

(speed, style, etc)

Attending TA hours!! (details TBA)

(13)

Any questions ?

You can find the course information at

◉ http://adl.miulab.tw

◉ [email protected]

◉ slido: #ADL2022

◉ YouTube: Vivian NTU MiuLab

Thanks!

13