kilgnad

81 ms·

391.
by kilgnad 4y ago
The memory of a poem is stored information but your vocal cord is NOT stored information. "*2" and "+1" are vocal cords. They operate on any number of inputs just like how your vocal cord operates on any number of nerve
392.
by kilgnad 4y ago
chatGPT is a reflection of those who trained it (not openAI but the actual ground level people hired to label answers). It generates bullshit because a lot of bullshit was approved during training.
393.
by kilgnad 4y ago
But it's not memorized. The "poem" in the case of the machine learning model is constructed from a general algorithm. Similar to how 2x+1 has no concept of 3. When you input a 1, it does a *2 operation and a +1 operation and
394.
by kilgnad 4y ago
Ok sure. This makes sense. Let's be clear so I don't get off track again. Your point is that an isomorphism means an encoding of the picture exists in the network and is therefore a copy. My point of contention is that it's n
395.
by kilgnad 4y ago
This doesn't make sense. F(X) itself is not (1,5). You have to apply F for it to work. Paint and paint brushes can be thought of as functions. You apply the right inputs (brush strokes) and you get an output that is a painting. Therefo
396.
by kilgnad 4y ago
>Because the dual of a structure is clearly not another representation of the structure. Isomorphism means nothing. /s But it's not an isomorphism. I have a function that has a range that covers every single number in existence
397.
by kilgnad 4y ago
It's technically sound though. Let me give you an example. y = f(x) = x + 2 This is what's memorized. But with that equation you can plug in a specific x. to get: 3 = f(1) = 1 + 2 The training set here would be (1,
398.
by kilgnad 4y ago
Machine learning is an extension of the concept of linear regression in 2-D space to 9999999-D space. They also make the line more then just a "line" the equation produces more of a best fit complex curve in multi-dimensional spac
399.
by kilgnad 4y ago
https://news.ycombinator.com/item?id=34613042
400.
by kilgnad 4y ago
Over-fitting is a known problem and deliberately avoided. This does not prevent the curve from intersecting an actual training set. I mean technically, if there are many copies of the same data set pair, THEN you can call it over-fitting. B
401.
by kilgnad 4y ago
How so? I have an equation that makes a number: y = 2x + 3 And I have (x,y) = (1,5) I'm saying the y = 2x + 3 is the ONLY thing stored in memory. The (1,5) is not stored anywhere. (1,5) is training data. y=2x+3 is the mod
402.
by kilgnad 4y ago
This makes perfect sense and is expected. But you have to know, none of the images are memorized. It's a coincidence. I can explain. Think of the analogy of linear regression. You have X,Y 2D space. X is the input (the English sentence
403.
by kilgnad 4y ago
There's definitely a distortion field around the F-35 both from the positive perspective and the negative. As most of us are laymen who've never flown the plane, it's hard to know which viewpoint is definitive. Sentiment has
404.
by kilgnad 4y ago
This isn't that dystopian. The dystopian outcome is when there's a classifier that rates the quality of the text and that this classifier becomes indistinguishable from the AI-generated classifier because AI generated text is begi
405.
by kilgnad 4y ago
Any reviews of these things. How good is it actually?
406.
by kilgnad 4y ago
Hah! Old news. You're a decade late. I don't think this is one of those ideas where some one is twisting some other concept into an app. I think this fits, big time. Typically forming a union involves someone stepping up and paint
407.
by kilgnad 4y ago
This type of thing is questionable. It can be viewed from the other angle. If software engineers aren't licensed why do lawyers and plumbers have to be? The pros of being licensed are obvious. The cons are less so. One thing licensing
408.
by kilgnad 4y ago
No, one scenario is slightly better because it gives you the choice to quit. In the other scenario there is no choice. Times constantly change you won't always be in a situation where you have the finances to just quit just because you
409.
by kilgnad 4y ago
Really? it was generated by chatGPT. Just kidding. Thanks!
410.
by kilgnad 4y ago
>It's a mixed bag... when your near the ceiling for a given area, it's harder to think about scaling back. A union doesn't necessarily have to negotiate maximum ceiling salaries. It doesn't have to negotiate salaries
411.
by kilgnad 4y ago
I mean which do you prefer. Being on a team with terrible engineers, or getting layed off. Take your pick. It's not really a binary choice either. With unions it just increases the probability of you being on a team of terrible enginee
412.
by kilgnad 4y ago
This is a YC owned site and it is against the interests of YC and all it's incubated startups. A lot of founders come here, so yes that group is likely mostly made up of founders.
413.
by kilgnad 4y ago
Actually HN is not bad. This is way worse on Blind.
414.
by kilgnad 4y ago
I wouldn't worry about that. There are literally too many people with your background for this too matter. I think there will be a form of credential-ism and it will continue to be similar to the tech interview.
415.
by kilgnad 4y ago
from what I read (and I didn't read super extensively) my app idea fits within this structure that's outlined by the NRLB.
416.
by kilgnad 4y ago
Yes. But does the law require this. Does the law require the union to have a leader and be operational at all times? Apologies, I'm honestly ignorant about the exact nature of the law here. But from the informal descriptions that I rea
417.
by kilgnad 4y ago
The app itself coordinates this. It guarantees only people with company emails can join and these people can only join with a single accocunt. Blind already does this and is quite popular among tech workers. I'm just extending the conc
418.
by kilgnad 4y ago
No the thing that has changed is the attacks. The upvotes I'm getting on that post indicates support. Maybe two or three years ago I'd have negative karma.
419.
by kilgnad 4y ago
On a side note I feel we can take Unionization into the tech world as a start up or an App. Imagine a blind-esque App that just allows everyone participating to vote anonymously for a union. Then AFTER a union is official then leaders and i
420.
by kilgnad 4y ago
Someone should invoke the Unionization defense.

More ›