Scribd
Upload
59 views2 pages

Project EE331 2019S

This document outlines the project for an Introduction to Machine Learning course. It details that students will (1) propose an interesting machine learning task and collect labeled data for it, (2) present their proposed task and data for selection, and (3) apply machine learning algorithms to the selected datasets and compete based on performance. Students will form groups of up to 3 people. The project is graded based on a proposal, presentation, and final report submitted by each group and individual.

Uploaded by

Thanh Nguyen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
59 views2 pages

Project EE331 2019S

This document outlines the project for an Introduction to Machine Learning course. It details that students will (1) propose an interesting machine learning task and collect labeled data for it, (2) present their proposed task and data for selection, and (3) apply machine learning algorithms to the selected datasets and compete based on performance. Students will form groups of up to 3 people. The project is graded based on a proposal, presentation, and final report submitted by each group and individual.

Uploaded by

Thanh Nguyen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Korea Advanced Institute of Science and Technology

School of Electrical Engineering

EE331 Introduction to Machine Learning


Spring 2019 Project Outline

Issued: March 13 2019,

Policy

This course will introduce many essential concepts and algorithms in machine learning through
lectures and homeworks. Students will learn that machine learning is data driven, and the task
performance depends on the quality of the dataset in terms of size and label. This project will provide
students the opportunity (1) to think about an interesting machine learning task that differentiates
itself from all other tasks that have been considered before (recommend students to reference Kaggle,
http:
www.kaggle.com), (2) to collect data and appropriately label it to perform the task (the more labeled
data, the more valuable your dataset becomes),(3) to apply benchmark algorithm to set the bar for
performance and finally (4) choose a dataset from a list of selected dataset and compete with others
in terms of performance. Students are encouraged to form a group for this project - the
size of the group should be no larger than 3.

The project will be conducted in the following manner.

1. Submit proposal for an interesting task(s) and collection of dataset


• Clearly explain the task and include why it might be interesting and/or important and
how the task differentiates itself with other known machine learning tasks.
• Discuss the dataset you are planning to provide in performing this task in machine learning
manner.
• Provide description (format, size, datasize, noise) of the dataset
• Discuss how you are going to label(s) the data.
• Discuss measure(s) of performance.
• Discuss benchmark algorithm that you are providing.
2. Presentation of task and label dataset for selection
• Each group will be asked to present its task(s) and the labeled dataset(s) it has collected.
• Presentation will be evaluated and about 4 datasets will be selected.
• Announcement of selected dataset and each student will choose a dataset from the selected
datasets to compete.
3. Apply various machine learning algorithms and compete for best performance
• Datasets will be made available to the students.
• Leader board will be setup for each task.
4. Submit final report discussing algorithm

1
Schedule & grading

Task Submission date Percentage


Form a group March 15 -
Project proposal March 27 40%
Presentation April 26 20%
Final video June 14,15 40%

1. Each group will submit 4-page task proposal in pdf format including a plan for data collection
The project proposal should include the following items.
(a) Clear description of the task with discussion on (1) how it might be interesting/important
and also (2) the uniqueness of the task.
(b) Detailed description of the data which you will be providing e.g. format, size, number of
data, label.
(c) Any prior knowledge required to understand task/data.
(d) Provide measure for performance. You may have multiple measures.
(e) Any algorithm that you are providing for reference.
(f) Introduce group members and their responsibilities.
2. For selection, each group will make a 10-minute presentation of the task and data. Each
member of the selected groups will receive 7% extra points on the overall final
grade. Members of selected groups must participate in task other than their own. Dataset
and task will be evaluated in terms of uniqueness, datasize and data quality. Selection will be
announced next day. Each group will choose a task from the selection and work on task as a
team. Leaderboard will be operated to inform student of best results.
3. Each student will turn in a 6-page final report describing the algorithm and experimental result
of the group. Student may individually design an algorithm separate from the group and report
its performance. The analysis of the experiment is expected to be different from student to
student regardless of the whether the student has submited an individual or group algorithm.
The final report should be emailed to ([email protected], [email protected] until June
15. Each group must submit video presentation of their algorithm using video recording soft-
ware described below.

You might also like