_dps

20 ms·

211.
by _dps 10y ago
In intellectual and creative work, the mind is not just hardware but also software and data. It's entirely plausible that the software gets better fast enough (with the right life experiences) to dominate the hardware getting worse wit
212.
by _dps 10y ago
It seems this has been submitted before, but I found it fascinating and it's a subject I've seen discussed with interesting results here before. Perhaps it will get some more activity going this time around :)
213.
Can we know what animals are thinking?(medium.economist.com)
2 points by _dps 10y ago|1 comments
214.
by _dps 10y ago
> However, I did practically nothing to earn that internship ... As someone who spends a lot of time in Europe, I find this exclusively-individualistic view of "earning advantage" to be too limiting, and characteristic of US id
215.
by _dps 10y ago
Respectfully, this is begging the question :) You're correct that neural networks don't encode information in the way current digital machines do (i.e. with explicit records/files/etc encoded in bytes). I would disagree
216.
by _dps 10y ago
> The knowledge (heuristics) are encoded in the structure, not in some kind of information. When you're talking about neural computation, distinguishing between an evolved network structure/architecture and "information&qu
217.
by _dps 10y ago
To add something, and to clarify (though I don't think you've made any mistake here, but many people misunderstand this subject): The 33%-each Nash Equilibrium is the mixed strategy Nash Equilibrium of the micro game (i.e. a sin
218.
by _dps 10y ago
[AI/ML researcher/engineer here] At best you can partly do this, and even that is debatable. Here's a simple example: say Google builds a linear regression for the incremental likelihood to click a social post based on the d
219.
by _dps 10y ago
I think this is meant to be a "hello world" level puzzle. You literally just click the cells and type to edit them. Are you sure you can't edit once you highlight a cell? If so that's probably a bug.
220.
by _dps 10y ago
First, thanks for all your work on Murmur and SMHasher! I wonder if you have any recommendations for a one-byte-at-a-time hash that survives SMHasher? I've been using Jenkins OAAT and haven't bumped into any problems but I'v
221.
by _dps 10y ago
> Then how come an unfamiliar piece of source code is easier to follow than a mathematics paper using unfamiliar notation? This is clearly subjective. I read lots of mathematics research that I find more accessible than even pretty munda
222.
by _dps 10y ago
> Uber on the other hand dispatches 100% of the work to their drivers This is flatly false. Uber publishes available fares in real time, drivers choose to accept them or not. This is the entire premise of Uber. You might as well say
223.
by _dps 11y ago
> but we're actually swimming in data ... I think this is a "looking under the lamppost for your keys because that's where the light is" situation. We hear about tons-of-data and deep learning because teams like Googl
224.
by _dps 11y ago
> It is deceptively hard [...] The set of abstractions required to do it right are unobvious and nontrivial ... I would be very interested to hear more about this aspect, if you would be willing to share. I'm always curious about th
225.
by _dps 11y ago
While the overrepresentation of the repeat offenders is a valid concern, I feel like this makes a big point out of a relatively small change in conclusion. Old Analysis: slightly over half were incarcerated again New Analysis: about o
226.
by _dps 11y ago
> but they can't do that while retaining staff and recruiting new staff effectively I wish I had a better way to ask this, but... can you actually provide any evidence this is true? It is directly counter to my ~10 years of experien
227.
by _dps 11y ago
I'm struck by the great historical fortune of the people in these images. Barring an interstellar collision, they will outlast any earthly monument we could create to honor a person. Long after the Earth has been consumed by the sun, J
228.
Photos NASA picked to explain our world to aliens(vox.com)
51 points by _dps 11y ago|8 comments
229.
by _dps 11y ago
The calibration cost is trivial compared to the coefficient learning cost. Very roughly, calibration is O(records) whereas coefficient learning is O(records * features). So the tiny add-on cost of calibration shouldn't affect anyone&#x
230.
by _dps 11y ago
> One thing that Naive Bayes sucks at is providing good probabilistic estimates. This is true based on how it's described in textbooks, but in practice it should always be combined with a calibration algorithm (which add a trivial O
231.
by _dps 11y ago
A person practicing "monomesitia" would be a "monomesitis". I agree "monomonger" sounds pretty vulgar, but I found it humorous :)
232.
by _dps 11y ago
I'm not familiar with the associated ancient Greek word for middleman (-poly and -psony come from ancient Greek), but in modern Greek "μεσαζοντας" (pronounced meh-sa-zon-das) is "middleman". So that would suggest so
233.
by _dps 11y ago
Sure, I elided a bit of complexity. I visit many coffee shops, including Starbucks. Some are owner-operated, some are not. Indeed Starbucks doesn't have a franchise model. And yet the managers both there and at owner-operated shops alw
234.
by _dps 11y ago
> a customer that sits half a day with a laptop costs Starbucks money. First, congratulations on your launch! I hope to try Wheelys out soon! A bit of well-intentioned feedback regarding your positioning strategy: as I am often the custo
235.
by _dps 11y ago
Their total value is irrelevant; what matters is the ROI you get from increasing or reducing each of them. I submit that taxes offer the lowest ROI of the three from the perspective of most businesses (but I admit that paying taxes to be in
236.
by _dps 11y ago
Are you saying something other than "you, the programmer, must declare spi to be volatile, or else the compiler might lay a bear trap for you?". Based on your comment it seems clear that you understand that... but given that you
237.
by _dps 11y ago
Part 1 discussed last week on HN: https://news.ycombinator.com/item?id=9724990 The comments here and on the blog raised a few interesting questions: - small route suboptimality problems, probably arising from integer input
238.
Solving the Traveling Tesla Salesman Problem: Part 2(mortada.net)
9 points by _dps 11y ago|2 comments
239.
by _dps 11y ago
Might you by chance be familiar with Rodney Brooks' work on subsumption architectures [1]? If not, I would summarize the underlying idea (my words not his) as "don't try to jump too many layers of abstraction in one go"
240.
by _dps 11y ago
That's an interesting result; do you happen to have a citation? I couldn't find one with a few minutes of Google Scholar. If I understand Concorde's claims correctly, there is still the question of finite numerical precision

More ›