![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
visit to EE dept
Apr. 3rd, 2003 02:24 amToday, Marcos introduced me to the EE department (that's what I was trying to do when I crashed the car on Dec 7). He's an undergrad doing research at the theory group.
I used to be prejudiced against EE's: I thought of them as the "circuit and signals people". I used to imagine EE as a messy science, always reworking the same old problems in more complex cases. But, as it turns out, it's full of elegant schemes and proofs, a lot of them involving analysis and number theory, and using mathematical principles of nature, such as "information theory". Applications such as error-correcting codes and cryptography use neat abstract algebra. Not to mention the cool epistemic flavor of cryptography research, and theoretical cybernetics research (multi-agent systems, control systems, planning, etc) which overlaps with Computer Science.
One of the interesting things about this kind of research is deciphering or inventing physical meaning. For example, Marcos says his advisor has meaningfully and usefully defined sine and cosine for finite fields. Part of the creative work is about applying old mathematical techniques in new domains.
Their building is noticeably less clean than Informatica's (which makes its own money through private partnerships), but the quality of the people doesn't seem to lag behind.
I watched a one-hour seminar about diadic representations of rational numbers (I never understood the point, but anyway) and a cryptography scheme by McEliece. It was temporarily entertaining to imagine the attacks being described, but it stopped being fun when I realized I didn't understand what was going on.
Tomorrow (i.e. today), there will be another one, probably about wavelets. I intend to check it out.
By the way, did I mention that college kids here have to pass much more stringent requirements than kids in American colleges? It's probably fair to say compare departments here with MIT as far as toughness goes. That's probably why there is such a low success rate. I think about 30% of students who enter university graduate on schedule. I'm talking about engineering in general. I should get some real stats though.
I used to be prejudiced against EE's: I thought of them as the "circuit and signals people". I used to imagine EE as a messy science, always reworking the same old problems in more complex cases. But, as it turns out, it's full of elegant schemes and proofs, a lot of them involving analysis and number theory, and using mathematical principles of nature, such as "information theory". Applications such as error-correcting codes and cryptography use neat abstract algebra. Not to mention the cool epistemic flavor of cryptography research, and theoretical cybernetics research (multi-agent systems, control systems, planning, etc) which overlaps with Computer Science.
One of the interesting things about this kind of research is deciphering or inventing physical meaning. For example, Marcos says his advisor has meaningfully and usefully defined sine and cosine for finite fields. Part of the creative work is about applying old mathematical techniques in new domains.
Their building is noticeably less clean than Informatica's (which makes its own money through private partnerships), but the quality of the people doesn't seem to lag behind.
I watched a one-hour seminar about diadic representations of rational numbers (I never understood the point, but anyway) and a cryptography scheme by McEliece. It was temporarily entertaining to imagine the attacks being described, but it stopped being fun when I realized I didn't understand what was going on.
Tomorrow (i.e. today), there will be another one, probably about wavelets. I intend to check it out.
By the way, did I mention that college kids here have to pass much more stringent requirements than kids in American colleges? It's probably fair to say compare departments here with MIT as far as toughness goes. That's probably why there is such a low success rate. I think about 30% of students who enter university graduate on schedule. I'm talking about engineering in general. I should get some real stats though.