Course talk:CPSC522/Ensemble Learning

Comments This is a good introduction to Ensemble learning and I got what it really is and what the three different representations are. I would have liked more detail in the document. However, this is a good explanation of what it is.

Nitpicks:

• Typo: See Decision Trees for and example of ensemble base learners ->> for “AN” example.
• Remove the “To Add” section

Grading Scheme

   * The topic is relevant for the course: 5
   * The writing is clear and the English is good: 5
   * The page is written at an appropriate level for CPSC 522 students (where the students have diverse backgrounds): 5
   * The formalism (definitions, mathematics) was well chosen to make the page easier to understand: 5
   * The abstract is a concise and clear summary: 5
   * There were appropriate (original) examples that helped make the topic clear: 5
   * There was appropriate use of (pseudo-) code: 0
   * It had a good coverage of representations, semantics, inference and learning (as appropriate for the topic): 4
   * It is correct: 5
   * It was neither too short nor too long for the topic: 3
   * It was an appropriate unit for a page (it shouldn't be split into different topics or merged with another page): 4
   * It links to appropriate other pages in the wiki: 5
   * The references and links to external pages are well chosen: 5
   * I would recommend this page to someone who wanted to find out about the topic: 3
   * This page should be highlighted as an exemplary page for others to emulate: 3

If I was grading it out of 20, I would give it: 17

Cool topic. Couple of points:

Base learners aren't necessarily weak alone... ensembles pretty much always improve performance, so a base learner is weak compared to an ensemble (maybe clarify that), but an ensemble of, say, SVMs can have pretty strong individual classifiers. Obviously the typical example of a decision tree in a random forest is pretty awful by itself.

I'd agree it builds on supervised learning, but can also be used for unsupervised tasks with some extra work.. perhaps worth mentioning somewhere.

One or two full examples/walkthroughs of ensemble classification might be useful. Generally expanding some of the definitions with a little more detail and high-level examples would be good. E.g. I'm wondering what kind of classifiers are typically used in a stacked ensemble, and whether you'd use the same classifiers as sub-models and for the meta-classifier.

Diagrams are good, maybe clarify what a "bootstrap sample" is - subset of the training data? Altered in some other way? Both?

  * The topic is relevant for the course: 5 yup
  * The writing is clear and the English is good: 5 maybe check some uses of commas
  * The page is written at an appropriate level for CPSC 522 students (where the students have diverse backgrounds): 5
  * The formalism (definitions, mathematics) was well chosen to make the page easier to understand: 3 - maybe add some? possibly not that useful though
  * The abstract is a concise and clear summary: 4
  * There were appropriate (original) examples that helped make the topic clear: 1 add some examples for the different cases
  * There was appropriate use of (pseudo-) code: 0 not really expecting any here
  * It had a good coverage of representations, semantics, inference and learning (as appropriate for the topic): 4 could definitely use expanding
  * It is correct: 5
  * It was neither too short nor too long for the topic: 3 I would've liked to see this a bit longer
  * It was an appropriate unit for a page (it shouldn't be split into different topics or merged with another page): 5 definitely. It's possible something like random forests deserves its own page, but this is still needed as an overview
  * It links to appropriate other pages in the wiki: 4 it's not clear which internal pages you should link to, if any. Wikipedia links are OK
  * The references and links to external pages are well chosen: 4
  * I would recommend this page to someone who wanted to find out about the topic: 4 good overview
  * This page should be highlighted as an exemplary page for others to emulate: 3

Overall ~16/20

* The topic is relevant for the course: 5
   * The writing is clear and the English is good: 5
   * The page is written at an appropriate level for CPSC 522 students (where the students have diverse backgrounds): 4 (more mathematical machine learning and tables can be added)
   * The formalism (definitions, mathematics) was well chosen to make the page easier to understand: 5
   * The abstract is a concise and clear summary: 5
   * There were appropriate (original) examples that helped make the topic clear: 5
   * There was appropriate use of (pseudo-) code:  3 (Although there wasn't pseudocode, given what you covered it wasn't really needed. Could be helpful if the page is expanded though.)
   * It had a good coverage of representations, semantics, inference and learning (as appropriate for the topic): 4
   * It is correct: 5
   * It was neither too short nor too long for the topic: 3
   * It was an appropriate unit for a page (it shouldn't be split into different topics or merged with another page):  4 (This page could probably be split up a bit into a few different pages, 
     each of the representative methods just be their own page. Although I do think the page works as a broad intro on its own as well.)
   * It links to appropriate other pages in the wiki: 5
   * The references and links to external pages are well chosen: 5
   * I would recommend this page to someone who wanted to find out about the topic: 4
   * This page should be highlighted as an exemplary page for others to emulate: 4
      

If I was grading it out of 20, I would give it: 17 :D