1. Team 9 – Modeling the Neo-Cortex
2. Should this project be considered for the Best Project award? No
3. Should this project be considered for the top 3 project awards? No
4. On a scale of 1 to 10, how would you rate the overall organization/clarity
of the project report? 7
5. On a scale of 1 to 10, how would you rate the overall project idea? 7
6. On a scale of 1 to 10, how would you rate the overall research
contribution of the project idea, methodology and/or results?  6

The project is a lot clearer than the original proposal. It seems like the project found something between the proposed idea and the actual research. The idea to use a group of properties to define objects and then categorize them through Hebb’s Rule to see how the information propagates up the chain is a really good use of the ideas presented in the class to figure out a better understanding of how the brain categorizes inputs. The output that was discovered had no way to be traced back down the chain of neurons but still had a small form of semblance to the original input. It was interesting to see the shape models that were built for the objects and their properties and how they correlated to the input and output data.

This project uses the ideas that were discussed in the earlier parts of this course with Jeff Hawkins book about the lone algorithm that runs the brain. This project covered the topics that were discussed in the book and tried to use the idea of wire together fire together concept for neural networks to see if the algorithm at a basic level has the capability to represent data that is alike. The idea of feeding this information in to a simulated situation is nice, but would be better if the data itself came from the robot. Maybe it would be nice to use the properties built during the Odd one out problems to see if this project could be made to be further useful.

What I liked best about this project was the simplicity of the input being propagated up the chain and using that information to determine if the project was successful. The ability to recreate the algorithm with a modified Hebbian algorithm to make the higher chains in the network more stable wa s a very good idea. The project had lots of examples with graphics to explain the results and that made it easier for me to read. 

What I liked least about this topic was its inability to actually identify a pattern in the output received from their program. This was one of the problems with the proposal in the first place. There was no information pertaining to what the expected results would be.

This project didn’t seem like it had an idea of what it was going to find. This isn’t really a problem but it could have led them to unusable results that had nothing in common. correspond to what was presented in the initial project proposal?

The good thing about this report is that it didn’t change that much from the proposed ideas.

There isn’t a good description of the individual properties that were used in the test input. For example what properties did each of the objects tested have in common and what did they have that were different. It was all represented in binary models but not in words that could be represented in such a way that a person reading it could understand. 

In the original paper there was not a specific relationship to the projects proposed work and what they would have classified as success. This was also the case in the final report. They did not mention whether or not they had been successful and didn’t write anywhere what improvements could be made to their project to categorize it as successful. 

This paper could be improved with some sort of reflection fro the authors on whether or not the goals were successfully attained. Also it would be better if the paper would have included more information on how they chose the properties of the objects they used as their test base. It would be a good idea to start with basic objects that a child has a strong understanding of such as the face of a parent or the mobile they have over their crib or a decal that a child might have on the wall of their room that they play in. this could allow for better properties to start with since the child might be much more familiar with these types of objects.