kylebgorman

11 ms·

61.
by kylebgorman 11y ago
I think the funding agencies would be a much better place, since they have a near-monopoly in many fields. And they decide "how much work is enough" for the money they hand out.
62.
by kylebgorman 11y ago
In the science of management there is the "two-factor" theory which holds that the factors which promote job satisfaction and job dissatisfaction are essentially separate, so, that, e.g., increasing job satisfaction does not neces
63.
by kylebgorman 11y ago
Unlikely. Cardinal numbers in Old English, long before the slightest chance of contact with Arabs, were virtually identical to the modern system with respect to order of terms. "For a similar situation in language, think of accents and
64.
by kylebgorman 11y ago
The German "decade flip" is restricted to the tens and unit places; otherwise the order is English-like, with larger terms leading. The cardinal number systems for most major languages lead with larger terms (as in English). I don
65.
by kylebgorman 11y ago
Could it be that they were damaged by the postal equipment? Envelopes with keys often undergo this fate too, unless you pad the key with something of similar thickness.
66.
by kylebgorman 11y ago
Yes but the "universal" sense is vanishingly rare outside of religious boilerplate and doesn't make sense in the context here.
67.
by kylebgorman 11y ago
I fail to see how word-based models are character-based models with manual feature extraction. Word boundaries are read directly from deterministically tokenized inputs. And, as I mentioned upthread, it has been known for about ten years, l
68.
by kylebgorman 11y ago
"Old-school" Markovian language models (the vast majority of what's being used in production today) are mostly word-based but for text applications with tons of data, high-order character models are competitive with word-base
69.
by kylebgorman 11y ago
"Empiricism" does not usually refer to taking measurements (which seems to be what you have in mind); it is a theory of epistemology which holds that knowledge is primarily experiential. (It just happens that many psychologists ar
70.
by kylebgorman 11y ago
try GNU Aspell.
71.
by kylebgorman 11y ago
FWIW, basic machine learning approaches to "sentence boundary detection" (as the task is called) get 199 out of 200 of these right (without using the "two space" clue), and have for a while. (e.g., http://sonn
72.
by kylebgorman 12y ago
Python is notable among its peers in that it has a well-defined spec that is not just a list of bugs in the reference interpreter. It is thus important to note that some of the non-reference interpreters do already have the properties reque
73.
by kylebgorman 12y ago
This post seems to conflate language and implementation. IMO, Python 3 the language has tons of improvements and no regressions. The grief on the internet about Python 3 makes me seriously wonder how many people who don't like Python 3
74.
by kylebgorman 12y ago
For comparability, most people use the Penn Treebank-III WSJ data. Sections 03-06 are test, the remaining sections are train/dev. Most methods are based on some sort of simple feature templates and machine learning, so they should gene
75.
by kylebgorman 12y ago
Self-promotion: I wrote an open-source sentence splitter tool that outperforms the state of the art on the "standard split". It is also very fast. http://sonny.cslu.ohsu.edu/~gormanky/blog/simpler-sentenc
76.
by kylebgorman 12y ago
My claim is that this idea that Japanese and/or Korean can be shown to be related to any other known language is very much a minority and/or fringe position, held by few mainstream historical linguists. (My only source is held b
77.
by kylebgorman 12y ago
While I don't have a survey in hand, my impression is that few historical linguists believe there to be a demonstrated (genealogical) relationship between Korean and Japanese: they are generally considered to be isolates (i.e., languag
78.
by kylebgorman 12y ago
In natural language processing, we can get close to state of the art performance on nearly every major task with a linear model; usually, the feature sets contain what are essentially conjunctions of features, but these are chosen by hand,
79.
by kylebgorman 12y ago
Your prediction that DNNs will replace much of the pipeline is very interesting to me, but I hypothesize that you're at least partially wrong. I predict that DNNs will impact early stages of the pipeline which operate on continuously v
80.
by kylebgorman 12y ago
There's no shortage of FOSS software for automatic speech recognition. Most academic research these days uses the Kaldi toolkit ( http://kaldi.sourceforge.net/about.html ), which has an Apache license. There are (at leas
81.
by kylebgorman 12y ago
Indeed. A strong link between genes and phoneme inventories is very unlikely, since as far as we know, all typically developing babies are born wired to the learn the phoneme inventory of any human language equally well. It is only later (s
82.
by kylebgorman 12y ago
I am always very leery of big-ticket science published in the Proceedings of the National Academy of Sciences (PNAS), as this paper is. Many papers (including this one) are not subject to peer review because they use what has been called th
83.
by kylebgorman 12y ago
We can deduce facts about how the Romans pronounced Latin sounds by: * descriptions of the sounds by the Roman grammarians * wordplay, onomatopoeia, spelling errors in graffiti, etc. * the way in which sounds from other languages (particula
84.
by kylebgorman 12y ago
Could we get a more informative title?
85.
by kylebgorman 12y ago
Telling somone to use Python 2 is a neat and tidy solution . What problems would this possibly cause? Python 3 is fine too.
86.
by kylebgorman 12y ago
I don't think mobile gaming should play a dominant role in teaching programming. (If I wanted to teach that, though, I wouldn't reach for Python, however.)
87.
by kylebgorman 12y ago
The Mayan rulers carefully documented their vicious wars of conquest (among other things). Some believe the Incas, with their "quipu" ( http://en.wikipedia.org/wiki/Quipu ) recording devices, were on their way
88.
by kylebgorman 12y ago
I think you want pig-pork, not pig-ham. (Ham is an Anglo-Saxon term.) IIRC, this "theory" first appears in dialogue in Sir Walter Scott's Ivanhoe (1820).
89.
by kylebgorman 12y ago
The absence of recursion would not actually "refute Chomsky" (whatever the hell that means; it's not like Chomsky only has one angle). "Universal grammar" just means "whatever linguists think all human language
90.
by kylebgorman 12y ago
I worry that this you may be neglecting base rate. I hardly even _know_ anyone who doesn't have a Mac laptop, regardless of what they do. I'm not sure the "non-Mac user" is even a coherent natural class anymore.

More ›