The fact that it is a hole is relevant to this discussion. The physics and safety of a large object passing through a relatively small hole are the entire thing in question.
2,524 karma · joined October 30, 2009
You are correct that the effects are much more dangerous than those of cannabis; that's why so many people in this thread are saying cannabis prohibition as had a negative effect in this particular way.
Because these substances are obviously and explicitly marketed as substitutes for cannabis. Cannabis is the relevant comparison. There's no agenda, just reality. You're being very dramatic calling this a "disgrace" when it's really very mundane.
For program synthesis applied to the "Atari" deep learning benchmark check out https://arxiv.org/abs/1806.05695
I didn't know they had a stake in this.
It sounds like you're going to reject proposals for generics on the grounds that "you could just cast values of type `interface{}`" unless someone can show you a situation where you can't do that. The problem is that the people who want this sort of feature don't think that's even a solution.
A "stack" is a stack of (vector, scalar) pairs (a vector and its "weight"). At each time step, the stack has three inputs: the vector to push, the weight to push it with, and the weight to pop off the stack. Its output is the top 1.0 weight of the stack. The stack's behavior on a time step is divided in to three parts:
1. First, that pop weight is removed from the stack. For example, if you want to remove 0.6 from a stack like [(a, 0.4), (b, 0.4)], you'd remove all of a leaving you with 0.2 to remove from b, so the final stack would be [(b, 0.2)].
2. The pushed vector is placed on the stack with the given push weight.
3. The top 1.0 of the stack is blended together and returned. For example, a stack like [(a, 0.4), (b, 0.4), (c, 0.8)] would take 0.4 of a, 0.4 of b, and 0.2 of c.
This process is differentiable, and you can combine this with the usual recurrent loops to make a recurrent stack machine.
> Instead of making predictions from a state that depends on the entire history, an autoregressive model directly predicts yt using only the k most recent inputs, xt−k+1,…,xt. This corresponds to a strong conditional independence assumption. In particular, a feed-forward model assumes the target only depends on the k most recent inputs. Google’s WaveNet[1] nicely illustrates this general principle.