- What will be solved?
- What will be achieved?
- Why is this research important?
- Scientific
- Application
- Society
- Background
- Literature review
- Limitation of existing work
- What will it not solve?
- Comparison between different approaches
- Possible ways to solve the problem
- Supporting arguments
- Limitations of our own research
- How will you know when you have succeeded
- Datasets / Metrics
- Proprietary datasets?
- Reproducability
- Baseline POC
- P1, P2, P3
- What will you do if your original plan fails?
- What are your reduced sets of problems?
-
Will the nature of research raise ethical concerns?
-
If yes, how ill you address these and seek ethics approval?
-
Promising areas: zero-shot learning, symbolic ML + DL, multi-task learning transferability
Choosing a topic for research is not easy. It is necessary to take into account many factors and risks, in addition, it is necessary to provide for the availability of material and the necessary literature. In order to deal with current problems and not waste time, I asked for help with research proposal from experts who work with the latest literature and always understand current problems and find ways to solve them. A research proposal is mandatory when applying for a PhD program, so I recommend taking it seriously. I was able to show the commission my competence, knowledge of the discipline, and desire to study the program and proved that my project deserves attention.