Hey guys. Can you confirm something for me please?
I'm under the impression that affirmative answers, in English, have rising pitch, whereas negative answers have falling pitch. Is this always the case?
e.g. "Do you play soccer?" --- "I do" vs. "I don't"
Consider the sentence:
You didn't leave the dog in the car, did you?
In spoken English, this statement may be spoken with a rising intonation or a falling one. If the former, it suggests that leaving the dog in the car is a bad thing, and might even suggest incredulity and consternation on ...
@ktm5124 Hmm... I think it can rise then fall or just, but neither I do or I don't normally keeps rising (and when it happens it would sound like I do? or I don't?), if I'm not mistaken.
@ktm5124 It doesn't seem like either of those refer to lying per-se. Equivocate seems closer though, since it seems more deceptive to use words with multiple interpretations and hope the wrong one is selected, than to just dodge answering a question.
in the meantime, I just saw this headline in my local neighborhood rag: " Stolen Apple Products are 'Easy' Money, Fedora-Clad Burglar Tells Police". Can someone be "hat-clad"? Doesn't seem right to me.
The officer issuing the permit is observant of the explosive level considerations, such as when detection of potentially dangerous atmosphere, in which the flash point of the product is higher than the ambient temperature, is not possible.
"The officer issuing the permit must ensure that explosive conditions, such as when when the flash point of the product exceeds the ambient temperature, cannot occur"
@Lawrence Various linear and logistic regressions, just wrapped up some neural network stuff
@Lawrence You know much about ML? I'm learning it because I just joined a new company and our product focuses on machine intelligence
thank you Dan, what do you think of the phrase "observant of ..considerations", is that anywhere close to sounding natural for saying that someone is mindful of rules
in this case we're using ANNs to help calculate things like probability of default, loss given default, risk weight assets, etc, all driving towards a more precise and reliable CCAR calc
@DanBron Heh :) . It turns out that single neurons are not general (see Minski). I exploited that to get a sqrt(n) factor improvement, if I remember correctly. Multilayer networks are more tricky, but I managed to distinguish relevant terms from less relevant ones, and got the more relevant ones out incrementally.
No one ever bothered to give me a PhD, those jerks
What language or platform did you use? For the sake of this coursework, I've been reintroduced to MATLAB, which I now remember why I both love and hate
@DanBron Sounds exciting. With the way banks are being forced to jack up their Tier 1 ratios or whatever, their ability to lend (and hence do business at all) must be under a lot of pressure.
@DanBron The back-prop algorithm was implemented in MATLAB, but my extraction algorithm (given a trained neuron or ANN) was implemented in Java (that's version 1, or close to it). The parallel programming constructs were really useful.