lars

22 ms·

151.
by lars 15y ago
Absolutely, there's a pretty clear mathematical justification for regularization. However, it is very literally tacked on at the end. Take logistic regression, if you minimize the cost function without regularization, you get a max-likeliho
152.
by lars 15y ago
A good example is regularization. You have nice proofs saying that your classifier is optimal, then you tack on a regularization term to it, which breaks your optimality proof but improves your classification accuracy. It seems unexpected,
153.
by lars 15y ago
The human learning to adapt to the machine has been the way EEG-based brain computer interfaces have been made for a couple of decades. Using machine learning to adapt the machine to the human is a much more recent development. It is possib
154.
by lars 15y ago
> We will wear a device which will be able to read our brainwaves and determine which word we are thinking ala dictation Since this thread is presumably being read by entrepreneurs making bets on the future of technology, it needs to be
155.
by lars 15y ago
No need to lose hope. There already was an (excellent) machine learning class last semester, so they should have the material ready.
156.
by lars 15y ago
If your argument is that Javascript isn't object oriented, you don't know Javascript. It is very much object oriented, it just uses prototypical rather than classical inheritance. If your argument is that a choice should be made between tea
157.
by lars 15y ago
It definitely isn't possible now, and I wouldn't expect it in five years either. If you look at [1], you can see the areas of the motor cortex. With todays methods we can do an acceptable job at separating for example hand from foot movemen
158.
by lars 15y ago
I'm not being clear, I don't think that thinking the words "left left up up" would be detectable through EEG. When I say detecting movement, I mean things like imagining moving a hand, a foot or a tongue. These movements use distinct areas
159.
by lars 15y ago
They're a little fuzzy, but the exact quote from the video is: "IBM scientists are researching how to link your devices, such as a computer or a smart phone. So you just need to think about calling someone, and it happens." From the context
160.
by lars 15y ago
Sure. Let's say a five second EEG segment, recorded from up to 200 electrodes. I bet that by 2017 we cannot accurately detect who a person would like to call out of phonebook of 50 people. Specifically, in a setting where each person is as
161.
by lars 15y ago
I've recently completed a masters thesis on EEG based mind reading, and I think I have a fairly good grasp on the state of the art in this field. I also have a copy of Kurzweil's The Singularity is Near by my bed, and I'm usually strongly o
162.
by lars 15y ago
That article was so good I made a separate submission for it. Thanks.
163.
If money doesn't make you happy, you probably aren't spending it right [pdf](dunn.psych.ubc.ca)
1 points by lars 15y ago|0 comments
164.
by lars 15y ago
Deflation creates an incentive to not spend money, because it will be worth more tomorrow. This kills the economy.
165.
by lars 15y ago
A 20-char password limit makes sense in one setting: When the password is stored as fixed length plaintext in a database. This is a really, really bad sign to see with a payment processor.
166.
by lars 15y ago
It's syntactic sugar for string concatenation. The same bugs would happen today if you were concatenating strings. I don't see the problem, honestly.
167.
by lars 15y ago
I'm doing a master thesis on EEG classification. Some thoughts: 1) The term Brain-Computer Interface is not made up for this article, it is the accepted name for this kind of thing in academia. 2) Their 85% accuracy is not classification ac
168.
by lars 15y ago
Could it be that you are in their Gmail contact list? I don't have a public Google profile, and never gave Facebook my Gmail password, yet I get suggestions for people I have emailed once years ago. The only explanation I can think of is th
169.
by lars 15y ago
There's a company doing that. Here's an article: http://www.freakonomics.com/2011/09/27/an-algorithm-that-can... “Only about 15% of what we do is traditional forecast meteorology,” says CEO Bill Kirk I once had a brief conversation with
170.
by lars 15y ago
They've also locked themselves into a resolution, in some sense. IPhone apps, unlike Android apps, are built using absolute coordinates. Developers can make assumptions about the width and height of the screen. And widths and heights are de
171.
by lars 15y ago
Someone listening in on your connection could take over your bank account session, pay pal session etc. It's unclear from the article how practical it is though, since it seems to require a JS injection on the attacked site.
172.
by lars 15y ago
If I remember correctly, it's mentioned in Artificial Intelligence, A Modern Approach that this problem is not only hard but in fact formally undecidable in the general case.
173.
TechCrunch controversy explained by Taiwanese CGI animation(youtube.com)
1 points by lars 15y ago|0 comments
174.
by lars 15y ago
I looked at the code, and what this does is to make calls to a web api on the errorjs.com domain. It sets a key using: http://rastajs.errorjs.com/set/yourkey/yourvalue And retrieves it using: http://rastajs.errorjs.com/get/yourkey It al
175.
by lars 15y ago
From a machine learning perspective this doesn't seem like such a daunting task. We have pretty good speech recognition for human language (plus acceptable speech synthesis). Assuming dolphins have a smaller vocabulary than humans, speech r
176.
by lars 15y ago
Leverage the tech savvy user base and start a start up incubator. Take a small part of each company's equity.
177.
by lars 15y ago
Whoops. Fixed it, thanks!
178.
Apple R&D (1989): Mousing is faster than keyboarding, but users don't believe it(asktog.com)
119 points by lars 15y ago|159 comments
179.
by lars 15y ago
> I think the main problem is computers are still too slow I don't think this is right. Depending on your neuron model, you'd probably be able to simulate as many neurons as is in the human brain if you threw it against, say, Google's s
180.
by lars 15y ago
And for automatic analogical reasoning, we've had the Structure Mapping Engine [1] since the 80's. I don't think focusing on what makes humans unique is necessarily the way to go, I think a better focus would be on what makes animals in gen

More ›