Deep Reinforcement Learning

CSC 261 - Artificial Intelligence - Weinman



Reading primary scientific literature is an important skill. Comprehending and assessing a work, even without understanding every detail, usually begins by garnering basic answers to the following questions.
Write one sentence in your Reading Journal to respond to each of these questions for the Nature paper by Mnih et al.
  1. What general problem/question are they trying to solve/answer?
  2. Why is the problem/question important?
  3. How is it currently done/answered (by previous work) and what are the limitations?
  4. What are the authors' goals?
  5. What is the paper's scientific thesis (if any)? Is it falsifiable?
  6. What are the paper's claims?
  7. Are the claims substantiated (by theory or experiment)? If so, how?
  8. What are the limitations of the proposed approach/answer?
  9. Are there ways to extend the method?
Acknowledgments  
The questions above are inspired by and adapted from the following works.
Fong, P. W.L. (2009) Reading a computer science research paper, SIGCSE Bulletin, 41(2), 138-140. doi:10.1145/1595453.1595493
Keshav, S. (2007) How to read a paper, SIGCOMM Computer Communication Review 37(3), 83-84. doi:10.1145/1273445.1273458.