Project Number 12:
Autonomous Learning in a Simulated Environment

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

2. Should this project be considered for the Best Project award? (yes/no)
NO

3. Should this project be considered for the top 3 project awards? (yes/no)
NO

4. On a scale of 1 to 10, how would you rate the overall organization/clarity
of the project report? (1-10)
8...The organization of the project was good.  It starts with a brief and terse introduction.  It goes into previous work nicely.  There are smooth transitions.  It doens't feel choppy.  However, related and previous work doesn't really seem to correspond with project.  It's too general.

5. On a scale of 1 to 10, how would you rate the overall project idea? (1-10)
10...This project idea is great and innovative.

6. On a scale of 1 to 10, how would you rate the overall research
contribution of the project idea, methodology and/or results?  (1-10)
8...I think this project is relevent to AI however since it's implemented via a simulation and not a embodied system I give it an 8.


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
* Overall, is the project report clear, concise, and well-organized?
The project is concise and well-organized.  I like that it includes a section that discusses changes made from the the previous proposal.  The previous work section could have used a little more "fluff".  It should have had more information about the literature that has inspired this research topic.  The description of the game is brief and to the point.  There is a nice introduction to reinforcement learning.  I would have liked to see a more "average-man's" definition for SVMs for a reader that isn't familar with this feild.  I don't really understand the structure of the program.  There should be a more in depth description of algorithms.


* How does the project idea and methodology fit within the framework of Developmental Robotics?
I think this project fits well with artificial intelligence and machine learning in general.  However I don't know if it directly relates to developmental robotics.


* Describe what you like BEST about the project?
What I like best about the project is the simulations ability to learn to avoid hitting certain areas of the wall after learning that it causes death.


* Describe what you like LEAST about the project?
It's implemented with a virtual similation, which is understandable as everyone could not work with an embodied system.  Also, I wish you couldn't gotton to a point where the simulation was actually fully engaged in the snake game.

* Do the methods, results and contributions of the final project
correspond to what was presented in the initial project proposal?
The results didn't really correspond with the initial proposal but the "change from proposal" section took care of this.  Also, I was looking forward to see the simulation really play the game fully in real time during the in class demos.


* Are there any major details left out with regards to the methods,
algorithms, or experimental design described in the report?
The methods, algorithms, and experiment is not really described describe clearly.  The structure is not well defined.  It could use charts or images to demonstrate how things come together.

* Do the experimental results reported in the paper demonstrate success?
The results demonstrate success although its not necessarily in line with the original proposal.

* Do you have any suggestions for improvement and future work?
Work on getting the simulation to play a complete game of snake.

* How close is the final project report to being publishable as a
conference or journal paper (consider the research papers that were
part of the course reading)? What would it take to get there?
I think this paper could be eligible for submission into a conference if it involved a developmental approach and if it involved an embodied system.