Course Overview
With the increasing complexity of data in recent years, classic machine learning algorithms have been difficult to meet actual needs. Learning algorithms adapted to different complex environments and different fields have continued to emerge, targeting complex data objects, complex tasks, and simulating human brain thinking. Machine learning theory has gradually become the mainstream of data mining and intelligent information processing. Mastering and proficiently applying these new machine learning theories and related algorithms has become one of the necessary skills for graduate students in computer science.
Course Webpage: https://www.aminer.cn/aml
Staff Information
Call For Proposals
- The goal of the course project is to design new ML algorithms for real problems, or to solve novel problems using ML algorithms.
- Work in teams of 1 or 2 people
- Schedule: proposal (6th week, Mar 25, 2020), mid-term (10th week), presentation (16th week)
- Grade: Proposal (15%) + Mid-term (15%) + Final presentation and report (70%)
Example tasks:
- Concronavirus Outbreak Prediction
- Author name disambiguation
- HotpotQA (https://hotpotqa.github.io/)
Open Tasks
- You can pick a project topic by yourself
- Data challenge
- https://www.kaggle.com/competitions
- https://biendata.com/
Provide a concrete proposal for a model or algorithm that potentially extends or improves the topics that you’ve picked.
- What is the problem you are solving?
- What data will you use (how will you get it)?
- What work do you plan to do the project?
- Which algorithms/techniques/models you plan to use?
- How will you test it?
Proposal submission
- A proposal in 1-2 pages in NeurIPS format, written in English (https://nips.cc/Conferences/2020/PaperInformation/StyleFiles)
- Deadline: March 23, 2020, at 11:59pm