Feedback on Project Proposal

[pooyanjamshidi]

I read your project proposal. Your project is clear in terms of scope. However, it looks like a very straightforward project, which is not bad, but I want you guys to learn more.

So, here are two alternative ways I would suggest considering:

• You can make this work an excellent empirical work by comparing all of those relevant techniques on the dataset, but you need to run many ablation analyses to produce some insightful results. Here I listed some examples of good empirical papers:

  • Cancellation in Systems: An Empirical Study of Task Cancellation Patterns and Failures
  • The Effects of Regularization and Data Augmentation are Class Dependent
  • MLaaS in the Wild
  • Measuring Robustness to Natural Distribution Shifts in Image Classification
  • Transfer Learning for Performance Modeling of Configurable Systems: An Exploratory Analysis
  • Statically Inferring Performance Properties of Software Configurations
  • Do vision transformers see like convolutional neural networks?
  • Pretrained transformers as universal computation engines
  • Evaluation of neural architectures trained with square loss vs cross-entropy in classification tasks
  • Model evaluation, model selection, and algorithm selection in machine learning
  • Multi-tenant GPU Clusters for Deep Learning Workloads: Analysis and Implications
  • Scale Efficiently: Insights from Pre-training and Fine-tuning Transformers
  • EfficientNetV2: Smaller Models and Faster Training
  • Why Do Better Loss Functions Lead to Less Transferable Features?
  • Do Better ImageNet Models Transfer Better?

• You can make your project a bit more challenging by integrating your diabetes prediction into diabetes detection and monitoring using some real devices. Many devices monitor biological signals in our body: Apple watches, Oura ring, whoop, etc. You can integrate your approach and show an indicator metric measuring the diabetes risk based on the person's biological signals live from the device.

[pooyanjamshidi]

Adding a bit more detail to my comments above, as you can see, there are even detailed tutorials about the end-to-end pipeline, from data preparation, to feature engineering, testing, and evaluation of different ML models, and visualizations of results. So, your empirical study should be more comprehensive than these tutorials.

I would like to give you another opportunity if you would like to define another project. If you would like to take this opportunity, please push another project proposal to this repository and tag it accordingly by September 16th, 2022.