hansvm

845 ms·

2851.
by hansvm 7y ago
Most modern browsers do.
2852.
by hansvm 7y ago
I think their point is that for most use cases, large images don't have a sizeable real world impact. They're more expensive to move around, and you can only fit so many large, distinct layers on a single machine, but a moderately
2853.
by hansvm 7y ago
I think the parent comment was referring to adtech in particular, which I agree isn't inherently bad, but which currently has a model of: 1. Slurp up as much data as possible no matter how sensitive and no matter the societal cost. 2.
2854.
by hansvm 7y ago
>integrable is quite a general class of functions (especially if you define the coefficients in terms of the lebesgue integral instead of the riemann integral). Quite true, but it's still just a subset of all functions, and you&#x27
2855.
by hansvm 7y ago
>you mean construct the periodic extension? Yes. A more common language felt appropriate given the broad background of the audience, and I also didn't handle any boundaries or edge cases in the functions whose existence I invoked --
2856.
by hansvm 7y ago
Often this is the case, and pi crops up plenty of other unexpected places, but counterexamples abound as well. Pick any function defined on [0, 1], and replicate it along the real line at unit intervals. You can force a description of it us
2857.
by hansvm 7y ago
Haha, that's the strategy my wife and I use. If we know we want separate plates we'll just not eat earlier and/or bring a box home. Some of the portion sizes are literally more than a day's worth of food.
2858.
by hansvm 7y ago
Many mainstream languages have separate cryptographically secure rngs. The standard built-in rngs have tons of flaws for crypto work. A plausible attack vector here is something akin to a chosen plaintext attack -- request a stream of passw
2859.
by hansvm 7y ago
This isn't a complete answer, but consider a series of data points describing a line. E.g., (0,0,0,0), (1,1,1,1), (2,2,2,2), .... It's possible to have lines even in any n-dimensional space, but the data is "high dimensional&
2860.
by hansvm 7y ago
That depends on your threat model. A patient attacker can store all of your encrypted messages till some future point where decryption is economically feasible. Since they already have the weakly encrypted data, encrypting it again doesn&#x
2861.
by hansvm 7y ago
The neural network counterpoint feels odd, but with respect to summations and integrals as leaky abstractions I think the author's point is that deploying them effectively as tools typically depends on a mountain of hierarchical knowle