Day XIX:

Today we had another Space Weather group meeting. Dr. Dube explained his work on trying to see if a neural network can make a correlation between spikes in separate data streams happening at different times. He explained the x-correlation technique and how every point in one set is multiplied by every point in the other set and is then normalized. This produces a huge spike in the output data. Dr. Glenn showed his work on image processing and how each quadrant of an image is in a particular state is is compared to a baseline or "quiet state". When a CME occurs, each quadrant show a spike in the different between the quiet state and the actual state. The problem is to see if a neural network can take in the data from the image processing technique and find a pattern. This would ultimately turn everything on its head and result in the prediction of, as opposed to reaction to, Coronal Mass Ejections.

Day XVIII:

Today I started out doing a lot of the usual things. I work until about 2:30ish when I was notified that the Remote Sensing group was going to have their collect today on short notice. They greatly need help to set everything up and prepare for the fly over. I went outside and set up targets. We finished three minutes before the plane flew over.

Day XVII;

Today I was going to help the Remote Sensing Group with another data collection session today, but after we got most of the targets set up the plan was postponed. I wasn't too mad because it is fun being outside helping out, at least when it isn't raining.

When I got back to my work I started reading some of the papers posted on the RIT twiki site. I read one paper that caught me completely off guard. At this point in time I feel as though I have a pretty firm grasp on the Neural Network concept but I guess people have come up with even crazier ideas. This paper introduced this idea that was almost a composition of Neural Networks in that one overarching Network orders for the training and learning of minion networks. The outputs from these subnetworks would then be weighted to produce a final output. It is like a Neural Network electoral college with certain helper networks getting more of a say in the final output than others. I think that the overall point of making a Neural Network that is composed of smaller networks (as opposed to just making a bigger network) is that new information can be added with having to retrain the whole entire network. It is very cool but very confusing.

Day XVI:

Today I spent a lot of time looking at how to use the MatLab neural network tool. It seemed pretty straight forward until I tried to load the input and target data. It kept on giving me a Matrix Dimension error. I guess I need to learn more about MatLab syntax and, well, just more about MatLab in general. Ultimately I want to be able to combine both the abilities of a Neural Network and the tools available in MatLab (i.e. regression analysis, graphical representations, etc..) to better see the abilities of a Neural Network as different things change.

I was also going to help the Remote Sensing group with its second collect today, but apparently there were problems and it has been postponed until tomorrow.

This morning for a daily intern meeting, we all observed how our eyes adjust themselves to different light settings. First we stayed in a dark room for about ten minutes, letting out eyes get used to the dark setting. Then we covered one eye and preceded out into the light for a little walking, letting the uncovered eye adjust back to the normal light setting. We then returned to dark room and opened out closed eyes to find that we could see out of that eye but not the other. I have done this before so the result wasn't surprising but it didn't matter because it was fun nevertheless.

Day XV:

Today was very eventful. From 9 - 12 I helped the Remote Sensing group prepare for the flyover by setting up different colored cloths/fabrics. I got to go up on the roof of the building and pin down 2m by 2m "blankets". It was pretty windy but also very cool (double meaning). There were a lot of different experiments being done. Some of these experiment were dealing exclusively with Hyper Spectral Imaging and the techniques of detecting different materials, while others were using LiDAR or other means to detect variations in heat. The plane that flew over was mounted with all of these sensors. The plan made multiple flybys in an interlacing pattern (back and forth) so that the whole area we were working was captured twice.

For lunch I met up with Dr. David Lawlor. I have know Dr. Lawlor for years and, since he is a professor here at RIT, we had made arrangements to get together. We talked about a lot of things ranging from my work here to his recent vacation to the Caribbean. It was very a enjoyable and informative meeting.

Day XIV:

Today we had another Space Weather group meeting to touch base with everyone and see what different avenues of research should be explored further. We talked a lot about neural networks and what things should be done next in the testing and training process that would better suit our needs. After listening to Dr. Dube and his methods of dealing with a neural network, I realized I have a few flaws in my own techniques. For one, the network size I was trying to train was not large enough and two, my method of training was not sufficient. Before I was giving the network one input and expecting it to give its approximation of a sine wave. I found that it is better to instead provide the network with multiple ordered inputs of the sine wave and try to train the network to guess the next output. It is a much different from my way of thinking.

Day XIII:

I was back to doing many of the usual things today. I did more work analyzing the demo network and debugging my own. There are a lot of errors I have to fix and things I have to try. I really want to make this work.

Day XII:

Today we went on a field trip to the University of Rochester to get a tour of the new biomedical/bioengineering building and labs. It was pretty cool. We got to see the work being done on artificial tissues from not only creation standpoint but also the testing/characterization standpoint. They were using ultrasonic waves and other techniques to view and atually form the tissues. On a side note, while were at the U of R the fire alarm went off and the building had to be evacuated. Apparently there was no fire so we eventually went back inside and finished the tour.

Day XI:

I did a lot more of the same kinds of things I have been doing with one exception, we had to unscramble a word. Given the letters "IPESNRUHTERIRANP" I unscrambled it to say "uniphaserprinter" which was not the right answer. Another good but incorrect unscramble attempt was "U R Happier Interns". That one was pretty funny. The final answer was apparently "intrapreneurship". Wikipedia defines this as "A person within a large corporation who takes direct responsibility for turning an idea into a profitable finished product through assertive risk-taking and innovation". I guess it has something to do with motivating creativity and innovation within a company.

My mom was on campus today so I got to go out to lunch with her. We went to this India restaurant off campus. It was delicious.

Day X:

I did more of the same things I have been doing today. I did more coding and debugging on my attempt at a neural network and did more work on analyzing the abilities of the demo network. Dr. Dube stopped by to see how I was doing. He said that once I get my attempt up and running I should try to have it approximate a sine wave. He also said that I should write up a brief analysis on the abilities of the demo network including its responses to rotation, scaling and both at the same time.

Coming up this week are more masters/Ph.D defenses (maybe some I can better understand) and a field trip to the U of R on Wednesday.

Day IX:

I got the opportunity to listen to a masters dissertation today. It was on something about the use of LiDAR and another sensing technique simultaneously to calculate the biomass of an area or identify certain species of trees. It was pretty complicated and I'm sure I didn't understand some of the main components of the presentation. There was a lot of statistical analysis and other things that I had never heard of before (what is an eigenvector?). Besides this gap in knowledge, the central point of the presentation was made clear, that being the combination of LiDAR and the other sensing technique (HSI maybe) produced either no gain in the ability to detect certain things or a slight gain as opposed to using HSI alone. It was pretty deep.

After this I went back to work and began debugging my network attempt. There are some problems to think about but hopefully nothing to serious.

Day VIII:

Today I finished the majority of the work on my attempt at making a neural network. I made things much clearer than in my previous attempts and avoided many of the pitfalls I had encountered earlier. I still have yet to test my work in its entirety, but I feel that when I encounter problems I will now be able to identify and correct my errors. The only part I still don't completely grasp is the backpropagation technique/algorithm used to "recalibrate" the system after each iteration of inputs and outputs. Since many Neural Networks have in them many layers between the input and output layers (lending to the term "Hidden Layer"), it is hard/impossible to tell at what point during the process something went wrong. To solve this problem, the error between the expected and actual output is distributed throughout the whole network. There is a lot of math on the subject that I don't really understand but I understand the goal of the method and I can proceed forward with my work.

On a side note, I bought a burger from the cafeteria/dinning area today. While much of the food on campus is expensive, I have to say this burger was worth the money I paid for it.

Day VII:

Today was very eventful. For the first half of the day we took a field trip to JML Optical Corporation. We got a behind the scenes tour of their manufacturing and design processes. It didn't realize all of the nuances and problems associated with the production of lenses, but even more impressive how precise the lenses were being made to the desire specifications. I learned a lot about the use of series of lenses in a system and how special coatings on lenses can drastically reduce the reflection tendencies of glass. JML also produces high quality windows, which go under many of the same precise processes and done to the lenses, except for the obvious fact that they are not curved.

Afterward we stopped and ate and finally returned to work. I worked for a little while but then most of the interns started a volleyball game in front of the building around 4:00ish. It was pretty low key and fun, but we did get some interesting looks from people walking by.

Day VI:

Today i did more of the same things I have been doing with the demo Neural Network, but decided to further my attempt at trying to emulate and make one for myself. My first attempt was poorly planned and I ended up going back over the code i had written many times. I basically failed at that attempt. So I stepped back and took a different approach. I implemented all of my ideas but when it came to testing and debugging, things became very convoluted. So withdrew once again and decided to plan everything out and try to avoid coding things that could become very complicated. Although I still have a feeling that they will.

On a different note, on the way back from lunch today I noticed an area called the RITreat. I thought it was pretty funny.

Day V:

First Monday!

Today was a lot like the others. I spent more time working with the demo network and trying to get it to recognize different orientations of images. It took a long time to load and gave me mixed results. While before I was under the impression that the demo network was not very good at recognizing the different orientations of images, it began to show some limited signs of life. It identified some of the images correctly while others, not so much. I need to do some further testing. I can't wait for Wednesday to go on the field trip and play some volleyball.

Day IV:

sorry, this is 3 days late

On Friday we had a meeting among the other people in the space weather group. We basically took a look at what everyone was doing and decide what direction our work should take. I am supposed to exploring the abilities of the demo network. Afterwards I went to the computer lab and continued researching CME dates. I then decided to test the demo network with different orientations of letters. I provided the network with letters in their original orientation, rotated 45 degrees, and rotates 90 degrees, and then tested it to see if it could recognize letters slanted at 30 and 60 degrees. The process of rotating the letters and then calibrating the network took up a lot of time and there is still testing to be done. My initial conclusion, however, is that the software does not recognize different orientations very well.

Day III:

I spent most of today sitting at my new location editing and resizing pictures of the sun so that they could be used to "train" or calibrate the demo Neural Network. This work did not turn out as planned. The software could easily analyze two pictures but anymore than that and it would stop. I could figure out what was going on. While the processing was taking place, I tried my hand at creating my own mock network. After installing java and its necessary components, I went forward and started designing the software. It is very complicated and nuanced. I really only coded the general format of the network but i hope to expand upon it and learn more as I delve deeper into this area of study.

For lunch I went to the Student Alumni Union found the game room. It was underground and connected to a cafeteria. Even though I didn't have enough time to play any games, I definitely want to go back there.

About an hour before I left, I went around to see what the other interns were working on. I found myself observing the Remote Sensing group using a special program on computers. They were using it to analyze images by combining different frequencies of light (I think). It was pretty cool.

Day 2:

I am starting to gain a better understanding of the the project I am working on. The problem of detecting Coronal Mass Ejections earlier and with greater accuracy is being solved by developing an Artificial Neural Network. By training the network to recognize certain patterns in data (i.e. in images), it can then infer the outcomes of similar data. The "learning" process of the network is critical in order to ensure the correct results. Ultimately the Neural Network will be able to take in information on the status of the Sun from multiple sources and predict what will happen, even though that exact combination of data may never have been encountered before. It is pretty complicated and I am still trying to make sense of everything.

Day 1:

Today was the first day of the of my summer internship. Running around campus on the scavenger hunt was fun but tiring in the 90+ degree heat. Our movie was the best...the DVD we create not only locked awesome but was short and to the point. I had to create a blog today on www.blogger.com which i am currently writing right now (well, not as you are reading it). I am looking forward for what this experience has to offer.