fritzo

25 ms·

301.
by fritzo 5y ago
> Robots have no feelings. Whoa hold up. Absolutely robots have feelings. What are feelings? They're signals warranting theory of mind and empathy. Even a fence gate has feelings, when you see it trying to close but it needs a l
302.
by fritzo 5y ago
Interesting! Can you help me imagine attack scenarios? All I can think of is: - The changeset is authored by a trusted committer but the committer's tools have been locally compromised. - The public tool itself (e.g. black) has been co
303.
by fritzo 5y ago
Great work Victor, it's so encouraging to see Asperti & Guerrini's research receive engineering attention! Anton Salikhmetov also had some nice implementations [1] a few years ago. As I understand, even Andrea Asperti didn&#x2
304.
by fritzo 5y ago
> biologically dead How can we balance microbial biodiversity against macroscopic zoological biodiversity? Certainly the Thames has been thriving with a diversity of microbial life. Are there measures of biodiversity that are less zoolog
305.
by fritzo 5y ago
Nina Simone had the highest form > I got my hair on my head > I got my brains, I got my ears > I got my eyes, I got my nose > I dot my mouth, I got my smile
306.
by fritzo 5y ago
Scientific publication: https://journals.ametsoc.org/view/journals/atot/38/1/jtech-d...
307.
by fritzo 6y ago
Maybe "ownership by a foundation" like Apache or Linux Foundation? Projects I've worked on for corporations have sometimes been donated to foundations for neutral ownership.
308.
by fritzo 6y ago
Exactly. I see a beaming and anxious engineer standing at the railing of a wharf, watching a cargo ship slowly and unstoppably glide away to a distant port. As the engineer's scarf flutters in the wind she looks down at her trembling h
309.
by fritzo 6y ago
Can anyone recommend articles that focus more on past examples of bad behavior, and that motivate each abstract principle with examples of behavior that principle aims to guide. This article and the articles linked seem very vague. Maybe so
310.
by fritzo 6y ago
Consider the two cases. If loot-boxes have no monetary value, then Epic won't be losing any monetary value in the settlement. If loot-boxes do have monetary value, then Epic should pay, and loot-boxes should be a valid form of payment.
311.
by fritzo 6y ago
Does gold coin use more gold than Argentina?
312.
by fritzo 6y ago
Agreed, blind testing is important for statistical correctness and code review is important to avoid adversarial backdoors like dieselgate.
313.
by fritzo 6y ago
That is a question for the judge or jury. The statistician's job is to provide a probability distribution, and to argue for the statistical correctness of that distribution based on similar inferences made on validation data.
314.
by fritzo 6y ago
Code review is not the way to validate statistical inferences. Even during internal review of statistical inference code, committers argue correctness through a combination of readable code, readable tests, and statistical validation on a c
315.
by fritzo 6y ago
Agreed, and with ML we'll need both open source and open training data.
316.
by fritzo 6y ago
Agreed, a super interesting book. From the Mamluk Sultanate to oil wars influencing WWII...
317.
by fritzo 6y ago
Does anyone know what those wire grids are in East Palo Alto, just west of the landfill? I can't tell if they're art or synthetic aperture radar. https://www.google.com/maps/@37.4464497,-122.1130144,156m/
318.
by fritzo 6y ago
Thanks, I hadn't heard of Rainbow's End!
319.
by fritzo 6y ago
Call to action for neural artists: We need deep fakes to generate vast amounts of plausible noise and thereby conceal our sensitive private signal. I feel like the only way our species will regain a sense of privacy is when we're con
320.
by fritzo 6y ago
I pursued this question in grad school [1], leaning towards symbolic logic: Suppose we want to learn a "natural" programming language. Training data would be example programs that we believe should be easy to express in any langua
321.
by fritzo 6y ago
I like the Bean Machine's declarative syntax because it makes it easy to implement interesting inference algorithms that treat different parts of a model independently. The syntax makes it easy to work around Python's lack of copy
322.
by fritzo 6y ago
paraphrasing to emphasize irony: "Science should be based on solid data: published, auditable, peer-reviewed numbers. Data is good, data is objective, data is truth. "Academic hiring is broken. We can't base academic hiring o
323.
by fritzo 6y ago
As I understand it is gravity's inverse square law that makes the effect strongest closet to the glacier and tapering off over a couple thousand kilometers.
324.
by fritzo 9y ago
Thanks for being honest :) > ...difficulty of finding out what the actual interesting bits are. > Uber's more detailed writeup ... is vague We indeed struggled to balance accessibility and depth that blog post. One place to go de
325.
by fritzo 9y ago
Deep Probabilistic Programming Dustin Tran, Matthew D. Hoffman, Rif A. Saurous, Eugene Brevdo, Kevin Murphy, David M. Blei https://arxiv.org/abs/1701.03757
326.
by fritzo 9y ago
Edward: like Edward, Pyro is a deep probabilistic programming language that focuses on variational inference but supports composable inference algorithms. Pyro aims to be more dynamic (by using PyTorch) and universal (allowing recursion). P
327.
by fritzo 9y ago
Author here. We're thinking of this as kind of a "numba for R", where Rcpp is the "cython for R". Any feedback is appreciated!
328.
Show HN: Rcjit – An experimental Jit Compiler for R(github.com)
7 points by fritzo 9y ago|1 comments
329.
by fritzo 9y ago
ZeroMQ is LGPL, not GPL. http://zeromq.org/area:licensing
330.
by fritzo 10y ago
What was your answer to the hiring manager? How about: "[buzzword] systems need lots of training data. To process that data quickly, systems must parallelize and distribute computation. Concurrent queues are a basic building block in p

More ›