This is an old revision of the document!
- After reading first three chapters:
- list the main parts/components/stuctures of the model.
- How is their creation dependant on each other?
- Thinking about the scripts (recipes):
- What is the main reason (the biggest advantage) of using scripts? What kind of information it brings? (Hint: page 2, page 8)
- Truth is - authors don't get the “knowledge” of scripts straightforward - how is that “knowledge” represented in model and which (two) ways are used to get it.
- In last paragraph of section 3 theres described method that enhances the robustness of the model (binarization of all association weights ). Answer one of following questions (choose one):
- Why it works? (⇒ Why it should work the best?)
- Have you any idea how to make it different way.
- In experiements part, which tools enhanced the tasks Atribute recognition and Composite activity classification “the most”. Why?