My workflow for a research in machine learning

This post is based on my experience when I was an intern at USTH ICT Lab. It's aimed for everyone to read.

I divide the research into 3 stages.

Stage 1: Preparation

1. Define in detail what the problem is.
2. Define performance metrics.
3. Find dataset and relevant datasets.
4. Read top research papers related to the problem as much as you can but not too deep. Read abstract, read model overview, read results.
5. Check their source code if there are.
6. Summarize all the ideas, all results and all hardware specification.

Stage 2: Experiments

1. Pick 2 or 3 research papers that you feel you can reproduce the results.
2. Read them in detail.
3. Reproduce results.
4. Brainstorm idea to what can be done to improve the result.
5. Carry out your experiments.
6. Summarize your work.

Stage 3: Report

1. Write a research paper
2. Publish your work.