tansey

54 ms·

211.
by tansey 15y ago
> Good grief, it's so lame it's almost trolling. Because it is trolling. HN is reading, upvoting, and commenting on an economics essay by someone who studied "Fine Arts" at the "School of Visual Arts" in NY and now works at "CYCLE Int
212.
by tansey 15y ago
I totally agree. I am simply predisposed. These days I like to think I'm a little wiser, thoughtful, and empathetic than I was as a kid, so I walk things through a little more in my mind. :)
213.
by tansey 15y ago
I'm not proud that I stole from the vending machine. I'm proud that I figured out how to steal from it. Fundamentally, I see no difference between being a 17 year old in a basement figuring out that if you send excessive amounts of data i
214.
by tansey 15y ago
Thanks! Keep working hard and you'll be just fine. We all feel that we need to play catch-up to someone. My boss had like 25 publications by the time he was my age, so I know the feeling. Like others have said, use it to your advantage as a
215.
by tansey 15y ago
I'm interested to understand why you think so. Most people in college at the time were signing up for poker sites without any affiliate reference, so they would get nothing. At least with Aces Up they got something back. My poker bot was ag
216.
The Greatest Hacks of My Life(wesleytansey.com)
183 points by tansey 15y ago|43 comments
217.
by tansey 15y ago
I would consider some sort of free version of this app, so people can join without risk. Maybe make a version that limits you to 3 friends for free, or unlimited friends for $1.99?
218.
by tansey 15y ago
> they're completely opaque black boxes with structure that only plausibly mimics the absolute simplest of biological neural networks I guess this depends on how advanced your toolkit is. If you're simply using feedforward nets with f
219.
by tansey 15y ago
> Virtually everything we call “AI” today is either a theatrical display of essentially scripted behavior (that’s how most game AI works), a massive database (such as Google Suggestions and expert systems) or a vague and decidedly unin
220.
by tansey 15y ago
Isn't that possibly a selection bias though? Isn't it possible that the YC brand is still trending up while the market is starting to correct? Or do you believe there is no such thing as a YC brand and every YC startup is judged independent
221.
by tansey 15y ago
A few points to remember: * Lots of YC companies fail. * Even more pivot from the idea they're working on now. * Competitors are not always a bad thing. They help validate your idea and push you to iterate faster. * Better-funded competitor
222.
by tansey 15y ago
Here's one study showing that complexity grows exponentially with the size of the program: http://www.cc.gatech.edu/sparc/Resources/readings/berry.pdf Here's a study showing addressing changed requirements or ﬁxing program defects require
223.
by tansey 15y ago
But people think that way because of the utility of the hand. Worthless hands are abstracted into the same category (junk) because they all have about the same value. A royal flush, however, has by definition a guaranteed value with no risk
224.
Ask HN: How does (or could) machine learning help you?
15 points by tansey 15y ago|5 comments
225.
by tansey 15y ago
I know this is the internet and everyone (including me) watches every video link to any event, but please seriously consider whether you really want to watch the links people are posting here. Effectively what happens is the plane is being
226.
by tansey 15y ago
That was the same question I had. It seems that CrimeSpotting measures police reports, which are skewed toward whatever crime the police are cracking down on.
227.
by tansey 15y ago
Pretty awesome that numpy and scipy are now supported by .NET. Does anyone know if the nltk is supported yet?
228.
Steve Wozniak Comments On Jobs Resignation Announcement (cultofmac.com)
3 points by tansey 15y ago|0 comments
229.
by tansey 15y ago
Actually, 3D games are a GREAT test-bed for AI research. Some of the best AI algorithm competitions in academia are possible solely because of the game industry's awesome killing simulators.
230.
Machine Learning as Art(wesleytansey.com)
7 points by tansey 15y ago|1 comments
231.
by tansey 15y ago
Probably want to put [video] in the title. Also, the title has nothing to do with the subject of the video. It's a good talk though, and worth watching.
232.
by tansey 15y ago
>Looks like you've re-invented the signal to noise ratio formula Well, I didn't, Sharpe did. :) However, the important distinction is the notion of a risk-free rate of return. In this case, it's (loosely) the 1 upvote you automaticall
233.
by tansey 15y ago
That's an interesting insight. I'm not sure that's true. Lots of people play games where their rankings shift down if they don't constantly play. In a lot of cases, this actually increases engagement. I think we would need to see some evide
234.
by tansey 15y ago
>Ideally, you should have some other behavioral metric in mind (say, mean comment quality, top comment quality, bottom comment quality, engagement, etc) That's precisely what my metric is intended to measure. The community derives qu
235.
by tansey 15y ago
Thanks! I tried to make it as hacker-friendly as possible, hence the matching pseudo-code for both formulas.
236.
How Karma Should Be Measured(nashcoding.com)
211 points by tansey 15y ago|67 comments
237.
by tansey 15y ago
Interesting. To summarize their methodology: - Take the top 20 hotels from trip advisor - Filter out all non-five star reviews, plus any non-english, excessively short, or first-time reviews. Then sample (via a log normal distribution on re
238.
Breed 3-D Printable Objects, No Technical Know-How Needed(newswise.com)
29 points by tansey 15y ago|11 comments
239.
Joel Spolsky - A Unified Theory of Internet Startups(plus.google.com)
99 points by tansey 15y ago|21 comments
240.
by tansey 15y ago
AppHarbor is awesome. The development process could not be any easier. You literally just create an MVC project and push it to your AppHarbor git repo and you're done. They also have support for SSL, multiple collaborators, and error loggin

More ›