 |
|
| |
|
Intrinsic Representation: Bootstrapping Symbols From Experience, by Stephen Larson
WRT: 35 minutes
The following is the MEng thesis on which the paper, with the same title, was based. The thesis includes a description of the algorithms uses in map adjustment, growing, clustering, and cross-domain cluster association.
Intrinsic Representation: Bootstrapping Symbols From Experience by Stephen Larson.
On a total of one side of one sheet of paper, using 10 pt type or larger, with standard interline spacing and margins, respond to all the following.
There is too much to read, particularly if you are interested in multiple fields. Accordingly, you decide to start a new journal named The MIT Alumni Reader's Electronic Digest, which you expect to go by the The Reader's Digest. The idea is to produce each month, in electronic form, in a section titled Steps toward Visions, a half-dozen or so one-page reviews of articles appearing in Science, Nature, and other top-drawer scientific journals. There is to be a section titled-Hero of the Month, in which a randomly chosen MIT professor identifies a few of his/her favorite classics for the treatment.
Curiously, Winston is the first professor randomly selected, and he selects the symbol bootstrapping paper. Your job is to write the synopsis, duly attending to VSN–C and detail. According to journal policy, you are also to offer your views on who should read the unabridged article and what they would get out of reading that beyond what they get from your summary.
Human interest asside: After completing his MEng, Larson went to work in Wall street, at Morgan Stanley, but two years later, in 2005, he went back to graduate school at UCSD, to take the next step toward his dream, centered on getting a PhD in a neuroscience program at UCSD.