Proposal Number: 12
Best Proposal Prize?  No
Clarity Rating: 7
Project Idea Rating: 8

----------------------------------

FEEDBACK:

Introduction
The introduction is extremely thorough.   It not only discusses what the project entails, but also gives some definitions for developmental robotics and why developmental robotics
is important or useful.  I like that it also discusses the “type” of artificial intelligence that people are most familiar with and why the developmental approach is better. 
You’ve done an excellent job in combining work done in AI from the past and present.  I like that it provides recent examples like “Watson” and “Emily Howell.”  When discussing the
“modular boxes”, I think it would be good to briefly discuss how this decision came to be.  For instances, why are you using this method?  Has it been done before and, if so,
where?  Also, I’m a little unclear of what the goal of this project.  Is it autonomous learning in a simulation or modeling learning abilities of a child?

Prior Work
Here as well you briefly discuss some previous work in the field but it is too general.  In the introduction that was okay but here you should only mention work that directly
relates to what you’re trying to accomplish.  I guess you could make two separate sections: one for a general description of work done in this area and the other for the work that
motivates this project.  It’s good that you discuss the different algorithms that you plan to implement and compare.  But, I think you should discuss more of why these algorithms
were chosen versus just describing them.

Approach
The game specifications are explained nicely.  I’m not confused about the premise of the game.  Reasons for altering some the game properties for the purpose of facilitating
learning of the environment make sense.   However, the algorithms are a little vague.  Perhaps this would be a great place to insert a diagram of the important steps in the
algorithm so the reader can have a visual understanding.  Also, here you’ve explained the purpose for using these algorithms which is good.  Adding sound to the game is always
good.  But you don’t explain why this is useful in a simulation in terms of adding modalities.  Are you putting in sound just in case time permits you to experiment with the real
robot?  Is audio useful in the simulation?  The interface is pretty sufficient.  The layout is great and that you are able to select different ANNs and compare without going back
into the code.

You should really discuss with each other very specific deliverables.  It seems that only Week two and three is dedicated to the simulation.  Will it only take two weeks to
construct your simulation and test it?  There are requirements for the simulation that you can count as deliverables.  I think this project can be done within the remaining time
in the semester but not if the deliverables aren’t nailed down first.

Evaluation
It seems that you have an idea of how the results will be evaluated and future work.

----------------------------------

MY CONCLUSION:

Overall, the proposal was well organized but could have been a little more descriptive in methods and previous work.  You should tersely state the research question.  It met the
proposal guidelines and it does relate to developmental robotics.  What I like most are the requirements for the user interface.  There isn’t anything about this idea that I do
not like.  It’s a fun topic!  Only concern that I have is that you understand the time it will take to construct the components of the application.  It seems that your goal is to
implement AND test within two weeks, which is fine as long as you are aware of the “deliverables”.  It is definitely doable in the remaining time.