Course talk:CPSC522/Deep Reinforcement Learning

I think this is a pretty interesting topic, so I'm looking forward to getting a chance to read through your article for the finished version. Although it isn't complete enough yet for me to give a critique or any in-depth feedback, I do think that what you have is a good start.

This page is incomplete. Based on what it has so far, it looks good.

Deep Reinforcement Learning

Comments

I kind of feel bad because we pretty much have the same topic, so it is hard to make proper suggestions how to differentiate the two articles. There is an interesting paper criticizing Rainbow and the Atari DQN in favour of genetic algorithms: https://arxiv.org/pdf/1712.06567.pdf

Maybe this could add some new aspects to the wiki page? The page is not finished so I can not properly criticize it yet. It is missing links and the formatting is off. However, I like the intro. It is much clearer than mine. Maybe a good outline is to explain DRL for Atari and the Rainbow approach and then contrast the Uber paper which talks about genetic algorithms? I had a similar approach with my page.

Marking Scheme

I a scale of 1 to 5, where 1 means "strongly disagree" and 5 means "strongly agree" please rate and comment on the following:

   The topic is relevant for the course. 5
   The writing is clear and the English is good. 4
   The page is written at an appropriate level for CPSC 522 students (where the students have diverse backgrounds). 4
   The formalism (definitions, mathematics) was well chosen to make the page easier to understand. ?
   The abstract is a concise and clear summary. ?
   There were appropriate (original) examples that helped make the topic clear. ?
   There was appropriate use of (pseudo-) code. ?
   It had a good coverage of representations, semantics, inference and learning (as appropriate for the topic). ?
   It is correct. ?
   It was neither too short nor too long for the topic. ?
   It was an appropriate unit for a page (it shouldn't be split into different topics or merged with another page).
   It links to appropriate other pages in the wiki. ?
   The references and links to external pages are well chosen. ?
   I would recommend this page to someone who wanted to find out about the topic. 1
   This page should be highlighted as an exemplary page for others to emulate. 1

If I was grading it out of 20, I would give it: 10

Thanks so much, Fabian. When I was trying to pick the papers, I did look at deep Q learning page but the papers were different. Unfortunately, I can't change the paper now. Thanks for your comments though. I'll try to take a look at the paper you suggested.