In-class notes from 10/28.
→ identify / refine research goal →
→ Develop and refine specific research questions →
→ Design and refine study / work to answer research questions →
→ carry out study / implementation and interpret results →
→ write and present products and documentation →
And always: read background material / gather background knowledge
But this process isn't linear. There should always be a constant presence of reading background material and developing and refining study. The entire process needs constant refinement and you make your own path as you go along.
Background reading. The purpose of this is:
We'll still do background reading at every stage in the process, but right now since our interest is very high-level, that will affect what papers we read.
Using Learning from Simulated and Unsupervised Images through Adversarial Training.
What are the key problems?
"However, learning from synthetic images can be problematic due to a gap between synthetic and real image distributions – synthetic data is often not realistic enough, leading the network to learn details only present in synthetic images and failing to generalize well on real images."