rahimiali

28 ms·

241.
How high can you go?(n.nfshost.com)
1 points by rahimiali 11y ago|0 comments
242.
A New Caption That Works for Every New Yorker Cartoon(theatlantic.com)
4 points by rahimiali 11y ago|0 comments
243.
Pitch Meetings: How Hollywood assesses creativity [pdf](leeds-faculty.colorado.edu)
1 points by rahimiali 11y ago|0 comments
244.
Stan: a probabilistic programming language(mc-stan.org)
136 points by rahimiali 11y ago|45 comments
245.
by rahimiali 11y ago
I'm biased of course, but this is one of the most amazing things happening in San Francisco right now.
246.
Bamboo in construction: an introduction (chap 5)(inbar.int)
1 points by rahimiali 12y ago|0 comments
247.
MiniZinc, the medium-level constraint modelling language(minizinc.org)
6 points by rahimiali 12y ago|0 comments
248.
Toomim Bros Bitcoin Mining Concern(toom.im)
9 points by rahimiali 12y ago|2 comments
249.
by rahimiali 12y ago
Avoiding the need for feature engineering is one purported advantage of deep learning. The reality is that shallow methods also work on raw data. All you need a large data set and a model whose capacity can grow with the size of the dataset
250.
by rahimiali 12y ago
Thankfully, between keyme and Keys Duplicated, only the latter let's you copy keys from a photo taken in the wild. It's a dubious honor that requires pretty complicated 3d analysis.
251.
by rahimiali 13y ago
Sweet! Thanks for the kind note.
252.
by rahimiali 13y ago
The US is an odd place for locks. Locks here pander mostly to convenience and cost, not security. So high security keys like those are rare here. One reason is the way we build houses. It's easy to break into our houses through their w
253.
by rahimiali 13y ago
it turns our you need a fancy laser cutter to do this. melting brass reflects infrared everywhere, which can destroy everything within sight. one solution is to use a different frequency, which costs more. i run keysduplicated.com, and we l
254.
by rahimiali 13y ago
it was undergrads. and there were six of them, to be fair.
255.
by rahimiali 13y ago
you've actually probably done this yourself. it's often called "featurization". for example, instead of applying a linear learner on vectors x in R^d, you apply it to vectors f(x), where f computes a bunch of features on
256.
by rahimiali 13y ago
Signal processing mysticism repeats itself every 20 years and has been fueled by tremendous hype since its debut 400 years ago: 1. Linear Regression (which, admittedly, was amazing) 2. Fourier Analysis (which is linear regression on orthono
257.
by rahimiali 13y ago
a good machine learning course might indeed cover 1940-1980s operations research (nonlinear optimization, linear/quadratic programming, dynamic programming), and statistics from 1970-1990s (graphical models, markov chain monte carlo me
258.
by rahimiali 13y ago
a low-pass filter? it attenuates high frequencies from the candidate pool?
259.
by rahimiali 13y ago
hi. i wrote the code that runs shloosl. this is a good point. i'm going to stew over this for a few days. if you're willing to chat, drop me a note at ali@shloosl.com?