benchmarkist

9 ms·

61.
by benchmarkist 2y ago
Great, it's always good to reach a clear resolution.
62.
by benchmarkist 2y ago
That's even better then. Just let the AI read the literature and write the optimal compiler.
63.
by benchmarkist 2y ago
Looks like a great use case for AI. Set up the logical specification and constraints and let the AI find the optimal sequence of SIMD operations to fulfill the requirements.
64.
by benchmarkist 2y ago
That's good to know.
65.
by benchmarkist 2y ago
I'm not making any value judgements. Intelligence is not necessary for replication.
66.
by benchmarkist 2y ago
On some days my cloth ears are more dry than on other days so the sounds are more or less muffled depending on ambient conditions.
67.
by benchmarkist 2y ago
Sounds like you agree that there is no problem.
68.
by benchmarkist 2y ago
Not really sure if serious or not but good try anyway. For the record I thought your response was very funny.
69.
by benchmarkist 2y ago
I haven't defined any problems or solutions but certain people do seem to have defined the problem like you have. What I said specifically was that whatever people perceive to be a problem with the global human population is self-corre
70.
by benchmarkist 2y ago
In any event, good luck with whatever plans you think will restore natural balance. I've made my points as clear as possible.
71.
by benchmarkist 2y ago
It's obvious once you think about what enables our current industrial activity: millions of years of biological growth and sedimentation. But it's all equivalent to solar radiation that was incident on the planet over millions of
72.
Information Theory: Obstacles to True Artificial Intelligence (2016)(blog.paulbohm.com)
3 points by benchmarkist 2y ago|0 comments
73.
by benchmarkist 2y ago
Nature is much too complex to fit into people's anthropocentric perspective. Either we dial back the production of industrial poisons or the natural world will continue withering from our consumption of natural resources. The fundament
74.
by benchmarkist 2y ago
Either they'll have children or they won't and the ones who do will replace the ones who don't. It's like I said, whatever people perceive to be a problem is not actually a problem because it corrects itself without any
75.
by benchmarkist 2y ago
Humans are animals and animals which do not replicate are selected out of the gene pool. There is nothing to argue here.
76.
by benchmarkist 2y ago
I am very optimistic. I don't remember where I read it but whatever remains after the current industrial civilization will not have access to the energy resources necessary for reindustrialization. Might have been Derrick Jensen but I
77.
by benchmarkist 2y ago
It's a tautology. Replicators which do not replicate do not persist in the environment and so are selected out of the pool of replicators. So whatever genes persist in the environment are tautologically the ones that managed to replica
78.
by benchmarkist 2y ago
It's a self-correcting problem. The people who don't have children select themselves out of the gene pool and are replaced by those who do have children.
79.
by benchmarkist 2y ago
Why is it a problem? It's not like people are somehow special. We eat and shit like every other mammal but without us the water and land would not be full of synthetic chemicals and plastics. Seems very obvious to me that human industr
80.
by benchmarkist 2y ago
Wishful thinking. You can revisit this comment in 2028 and decide whether you were right or not.
81.
Significant amount of methane escaping East Siberian Arctic Shelf (2013)(phys.org)
3 points by benchmarkist 2y ago|0 comments
82.
by benchmarkist 2y ago
Doesn't seem like testers would have access to the actual weights/graph and so would not be able to leak anything.
83.
by benchmarkist 2y ago
Profits and subsidies are privatized, losses are socialized. That's how the system is designed to work.
84.
Beating GPT-4 with open source(blog.dottxt.co)
5 points by benchmarkist 2y ago|0 comments
85.
OpenAI CEO Sam Altman on AI governance, ethics, and innovation [video](youtube.com)
1 points by benchmarkist 2y ago|0 comments
86.
by benchmarkist 2y ago
You'd have to define what you mean by luck which would be equivalent to choosing some pattern that your compressor can recognize in a random stream of bytes and then computing the probability that pattern appeared in some random byte s
87.
by benchmarkist 2y ago
Take random bytes from /dev/urandom, encrypt it. There is no way any compressor can reduce the size of that file. I'm pretty sure there is no way the second player can win this game. Actually, the encryption part isn't e
88.
Evaluating the Robustness of Analogical Reasoning in Large Language Models(arxiv.org)
1 points by benchmarkist 2y ago|0 comments
89.
by benchmarkist 2y ago
Which makes it entirely irrelevant as a descriptive term.
90.
by benchmarkist 2y ago
Similar to sledgehammer in Isabelle/HOL.

More ›