darkmighty

13 ms·

61.
by darkmighty 7y ago
> no AI that I know of has at its disposal a full blown model of the world it operates in This is a field called Model-based Reinforcement learning, and it's quite advanced already -- there are indeed models that have an internal st
62.
by darkmighty 7y ago
I feel that reflects quite well Steve Jobs vs Tim Cook as well. Jobs just wanted to make 'insanely great' products [for their users], with the side effect of profit, while Tim Cook is probably fixated on efficiency and Shareholder
63.
by darkmighty 7y ago
I think the main danger is when truth and improvement is compromised in favor of niceness. Sometimes you are wrong, and that's definitely not a comfortable position to be in. It's not going to feel good to have it pointed out. And
64.
by darkmighty 7y ago
This appears to confirm a prediction of mine that black holes aren't really black holes -- you can escape them just fine ! I think physicists will become more and more convinced of this fact -- there can be no impassable communication
65.
by darkmighty 7y ago
If anyone is interested it seems you can find (more or less reliable) sensor modules for about $20, such as the MH-Z19, and read it from to an arduino or esp.
66.
by darkmighty 7y ago
One could call it a feeling of responsibility maybe. The ongoing conflicts were defining the world and scientists (and he individually) could change the course of history. I think vNM's work is equally impressive (or even more; it&#x27
67.
by darkmighty 7y ago
Most don't require cooling, but will have better clocks. The 4 itself throttles under heat (above 60C and heavily over 80C), and seems to have good enough behavior even if not actively cooled.
68.
by darkmighty 7y ago
This seems conflicting with what I read about its behavior. It should heavily throttle (underclock) itself, but shouldn't destroy itself. Maybe you got a defective unit or the hot thermal environments caused issues elsewhere (again may
69.
by darkmighty 7y ago
There are different models of computation. You can have a model in which integers can be arbitrarily large yet still count as constant memory (for analyzing practical-size algorithms on 64-bit machines, for example); I'm not sure how c
70.
by darkmighty 7y ago
I too like the idea of Fusion reactors, (I am enamored with the dreamy prospect of fusion for space travel), but I think Elon Musk's observation is obligatory here: Why spend huge sums in a fusion reactor, when we have a perfectly reli
71.
by darkmighty 7y ago
I get what you're implying, and I think it's a valid approach (connected). But so is the converse (unconnected). The connected can be controlled by apps or heck, APIs, perhaps offers full flexibility etc. However, what guarantees
72.
by darkmighty 7y ago
So I wonder if society (business, education, organizations) operates on the same schedule in Paris and Berlin? (or between Madrid and Budapest) I suspect there's at least some adjustment, which indeed could negate health effects even i
73.
by darkmighty 7y ago
Instruction complexities can be used to save bandwidth/delay (and related energy consumption) at the cost of decoder size. Communication limitations are increasingly dominant in processors afaik; so the analysis is not so simple as to
74.
by darkmighty 7y ago
Physical interaction is still extremely important to societies and individuals; not everything can be virtualized or made remote without loss of effectiveness or efficiency, at least not yet. For personal interactions, you have face-to-face
75.
by darkmighty 7y ago
Perhaps you could try CPU training. I believe it should take about 10x longer on a many-core machine, but still might be worth it (if you require lots of RAM and can afford to wait). There are also various techniques to reduce memory usage.
76.
by darkmighty 7y ago
Very nice. If I may ask for clarification on the CNN method (which seems to work quite well), you're taking a CNN that's built for classification but removing the last layer (which I believe is fully connected?) so that you only t
77.
by darkmighty 7y ago
I think most people don't imagine properly the world as automation increases at all levels. The first expectation is perhaps that mass unemployment would abide; clearly we don't have mass unemployment in most places (most of the w
78.
by darkmighty 7y ago
I did not read the paper fully, just some general commentary, hopefully useful. First, fantastic that you're doing academic work so early. I believe most students wait far too long to be exposed to this aspect of academia (and their li
79.
by darkmighty 7y ago
Being a good Bayesian would entail taking the evidence as-is, only adjusting your estimate of the "true" value after the fact taking other measurements into account.
80.
by darkmighty 7y ago
https://www.scottaaronson.com/blog/?p=4317 Scott's Quantum Supremacy FAQ
81.
by darkmighty 7y ago
It's very delicate because from personal experience most projects are likely to fail, and most projects start way over ambitious (almost a natural bias). What takes lots of experience to learn is also that that's not necessarily
82.
by darkmighty 7y ago
I think that might not be approaching it form the best angle? Perhaps the best question is: how to motivate students to excel academically, and give them opportunity to study to match their motivation? (instead of having poorly performing s
83.
by darkmighty 7y ago
Exactly. Shannon could have called his metric a 'Logarithmic measure on probability spaces', but he chose to call it 'Information' -- even though it's not exactly what we mean when we talk about information informal
84.
by darkmighty 7y ago
Birds have been in significant and will probably continue to decline due to human activity. https://en.wikipedia.org/wiki/Bird_extinction The main drivers I think are plain habitat loss, pesticide use (associated decli
85.
by darkmighty 7y ago
If reflection was diffusive it would be more or less trivial. If the room is connected, the is a path from the source A to any chosen point B (by definition). Now take this path and modify it such that it is composed of straight lines, and
86.
by darkmighty 7y ago
Even an infinitely conductive sphere does not screen the quasi-static near field very well. Although the delay is probably going to be weird, greater than direct visibility. Atmospheric EM events indeed are what comes to mind.
87.
by darkmighty 7y ago
Depends on how low the frequency is :) But indeed I'm not an expert by any means on magma properties, perhaps someone else can chime in. If magma is conductive then even less radiation goes through. Keep in mind at extremely low freque
88.
by darkmighty 7y ago
> no signal of terrestrial origin will travel at anywhere close to the speed of light through the Earth (earthquake waves travel orders of magnitude slower) To be fair, electromagnetic (i.e. radio) waves do travel at the speed of light t
89.
by darkmighty 7y ago
The real world is expensive. Everyone has a smartphone in their pockets (and most a moderately powerful PC of some kind depending on region), where hundreds of hours of challenge and entertainment can be had for free (or a few dollars). A p
90.
by darkmighty 7y ago
Is the relevant model here simply a sum of 2nd order terms? It's widely known to practitioners vanilla Neural Nets don't do very well on small datasets because of overfitting issues -- which can probably be corrected with very car

More ›