Monday, July 18, 2005

The Past 2 Days

Monday- We did more number cruching this morning. Figuring out some stuff with Google, like how much computational power it would need, how fast would x computers need to operate to do y amount of work within this time, etc. He seems to like doing these estimation things a lot, which is going to be a good skill to have in the future... although practicing it can be quite boring! We went over some sorting algorithms. It was actually very interesting, even though I learned it in computer science. AP ComSci mainly dealt with one computer, and thus was very limited in what it could do in terms of time. But COSMOS introduced the concept of parallel and distrubuted processing, which is quite interesting. We never discussed dividing work among multiple computers. Running multiple outer loops of bubble sort while the inner loops were still going was really, really cool... Quicksort was also pretty cool, because he talked about ideas for making the program faster, not just more efficient. Rather than waiting for each group to finish sorting, you can immediately begin the next step while the previous step is still working. Pretty cool concepts that we haven't touched in AP. In the afternoon, Christine discussed human and machine perception... like some difficulties in how to interpret images. We talked about some general guidelines to determine the correct interpretation called the Gestalt Rules. We worked in the lab a bit... some web activities dealing with perception, then some introductory C# stuff. I think Christine is much more organized in the lab. There's definite stuff to work on, rather than just go to the lab and do whatever. I think it lets us learn better, which is important because there's only 3 weeks left. Unfortunately, we still don't know more about the final project... except that it could involve programming. I think we need some definite ideas about what we can do, because everyone is still clueless about what it involves. This morning, we had a lecture on Brain-based robots... robots that could learn. There was one that could learn to avoid "bad-tasting" (non-conductive) objects and go to the "good-tasting" conductive objects and associate them with both images and sounds. But the really interesting one was a robot called Brainworks... it was an attempt to get robots to play soccer! It was really interesting, especially because the speaker had a lot of videos to show of the robot learning and performing. Very interesting lecture. We're going to go to the SuperComputer Center this afternoon! Should be a lot of fun... hopefully it won't be our only field trip.

1 Comments:

Anonymous Anonymous said...

LED flowering grow lights will produce light in the right frequency and wavelength.

They need only ten to thirty percent as much power as C7s or C9s.
The LED light provides more light and less energy use.


Feel free to visit my web blog - LED Stehleuchten

11:20 AM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home