erwald

8 ms·

61.
Midihum: An ML-Based MIDI Humanizing Tool(erichgrunewald.com)
55 points by erwald 3y ago|18 comments
62.
Attention on Existential Risk from AI Likely Doesn't Distract from Current Harms(erichgrunewald.com)
1 points by erwald 3y ago|0 comments
63.
10 Years of Earning to Give(forum.effectivealtruism.org)
2 points by erwald 3y ago|0 comments
64.
A List of (Almost) Every Single Chinese Tofu(brokencuisine.com)
3 points by erwald 3y ago|0 comments
65.
by erwald 3y ago
"Sure, you could prepare for imagined eventualities, or you could do the actual work of improving efficiency, reducing waste and unnecessary middle-men, and removing centuries old bureaucracies that are now absurdly pointless in the fa
66.
by erwald 3y ago
"Grok" in AI doesn't quite describe generalization, it's more specific that that. It's more like "delayed and fairly sudden generalization" or something like that. There was some discussion of this in the
67.
by erwald 3y ago
Sure it's easy -- you can use benchmarks like HumanEval, which Stability did. They just didn't compare to Codex or GPT-4. Of course such benchmarks don't capture all aspects of an LLM's capabilities, but they're a l
68.
A Nordic revolt against 'ugly' modern architecture(bloomberg.com)
91 points by erwald 3y ago|90 comments
69.
Tentative practical tips for using chatbots in research(forum.effectivealtruism.org)
1 points by erwald 3y ago|0 comments
70.
Alpha(erichgrunewald.com)
1 points by erwald 3y ago|0 comments
71.
by erwald 3y ago
The author of this article is Dan Hendrycks, who has a PhD in ML from UC Berkeley and was one of the contributors to the GELU activation function (among other things).
72.
by erwald 3y ago
That's weird, are you looking only at the top 10 countries? I've reproduced dwaltrib's results using World Bank data on 251 countries, and I get a Pearson's r of 0.82 and a p value of 5.6e-61 (!). I.e. a strong correlati
73.
by erwald 3y ago
> But this is obviously not true. Just look at the list of top 10 by GDP and check out how many of them are relatively small. There absolutely is a correlation between land mass and nominal GDP.
74.
Misaligned AI constitutes a growing public health threat(bmj.com)
2 points by erwald 3y ago|1 comments
75.
Yes, avoiding extinction from AI *is* an urgent priority(soroushjp.com)
2 points by erwald 3y ago|5 comments
76.
by erwald 4y ago
I'm curious, did you read beyond the summary? (I don't mean that in a snide way, it's totally fine just to read the summary -- that's why it's there.) The 1e35 FLOP number is meant as a conservative upper bound and
77.
The Prospect of an AI Winter(erichgrunewald.com)
144 points by erwald 4y ago|256 comments
78.
Power Smoothie(powersmoothie.org)
3 points by erwald 4y ago|0 comments
79.
by erwald 4y ago
I think I mostly agree with you, but I think this framing is a bit misleading. On autocomplete, I'll just lazily quote the relevant part from my post: It’s completely true that LLMs are trained on next-token prediction (although some,
80.
by erwald 4y ago
They definitely hallucinate a lot too. But they also seem to do things genuinely like reasoning, e.g. they seem to contain and use "cognitive" world models, and seem to be capable of learning fully general algorithms. Beyond that,
81.
by erwald 4y ago
Excess mortality still looks substantially higher in Sweden than in Norway and Denmark (though not Finland) in this OWID data set, I wonder why the difference? - https://ourworldindata.org/grapher/excess-deaths-cumulati
82.
by erwald 4y ago
> Aaronson’s biggest error, now corrected, sort of, in bold, is in assuming that Chomsky has spent his life in some sort of failed effort to build AI, which kind of entirely misses the point of Chomsky’s piece (which says in so many word
83.
by erwald 4y ago
IMO Chomsky (and Marcus) are mostly wrong on LLMs. Quoting the summary of a post[1] I wrote recently: - LLMs are sometimes said to be “just” shallow pattern matchers, “just” massive look-up tables or “just” autocomplete engines. These compa
84.
Against LLM Reductionism(erichgrunewald.com)
2 points by erwald 4y ago|1 comments
85.
How to Get Hold of a Research Paper(erichgrunewald.com)
13 points by erwald 4y ago|2 comments
86.
Notes on Meta's Diplomacy-Playing AI(erichgrunewald.com)
1 points by erwald 4y ago|0 comments
87.
Measuring Good Better(forum.effectivealtruism.org)
2 points by erwald 4y ago|0 comments
88.
Sweet Irony(erichgrunewald.com)
1 points by erwald 4y ago|0 comments
89.
How many effective altruist billionaires five years from now?(erichgrunewald.com)
1 points by erwald 4y ago|0 comments
90.
Emacs Lisp Functions for Easy BibTeX Recording(erichgrunewald.com)
17 points by erwald 4y ago|1 comments

More ›