goggy_googy

3 ms·

1.
Apollo 17 Descent Orbit Insertion(nasa.gov)
1 points by goggy_googy 2y ago|0 comments
2.
by goggy_googy 2y ago
What makes this such a "deeply broken situation"? I agree that late-stage capitalism can create really tough situations for poor families trying to afford drugs. At the same time, I don't know any other incentive structure th
3.
by goggy_googy 2y ago
I think at some point, we will be able to produce models that are able to pass data into a target model and observe its activations and outputs and put together some interpretable pattern or loose set of rules that govern the input-output r
4.
by goggy_googy 2y ago
Agreed. At the very least, models of this nature let us iterate/filter our theories a little bit more quickly.
5.
by goggy_googy 2y ago
No, the activations are a combination of the basis function and the spline function. It's a little unclear to me still how the grid works, but it seems like this shouldn't suffer anymore than a generic relu MLP.
6.
by goggy_googy 2y ago
I think the hand running through the wheat (?) is pretty good, object permanence is pretty reasonable especially considering the GAN architecture. GANs are good at grounded generation--this is why the original GigaGAN paper is still in use
7.
by goggy_googy 3y ago
Reminds me a little bit of a bloom filter in its functionality
8.
by goggy_googy 3y ago
This paper reminds me of the Neural Network Diffusion paper which was on the front page of HN yesterday in the sense that we are training another model to bypass a number of iterative steps (in the previous paper, those were SGD steps, in t
9.
by goggy_googy 3y ago
"We synthesize 100 novel parameters by feeding random noise into the latent diffusion model and the trained decoder." Cool that patterns exist at this level, but also, 100 params means we have a long way to go before this process
10.
by goggy_googy 3y ago
Important to note, they say "From these generated models, we select the one with the best performance on the training set." Definitely potential for bias here.