jal278

6 ms·

1.
by jal278 4d ago
appears it's just a theory, sort of surprised this lifted up so far on hn
2.
by jal278 4d ago
not sure w/o context why this is important -- no results/implementation & i believe there are prior combinations of transformers/RNNs. but perhaps I'm missing the relevance/insight
3.
by jal278 10d ago
> Teaching machines to love Reminds me of a research paper I wrote a few years back: https://arxiv.org/abs/2302.09248
4.
Intentionally Designing the Future of AI(goodfire.ai)
1 points by jal278 7mo ago|0 comments
5.
The MUD Client Protocol (MCP)(moo.mud.org)
2 points by jal278 10mo ago|0 comments
6.
by jal278 11mo ago
The fundamental premise of this paper seems flawed -- take a measure specifically designed for the nuances of how human performance on a benchmark correlates with intelligence in the real world, and then pretend as if it makes sense to judg
7.
by jal278 11mo ago
The function of news is to help a democratic citizenry be critically informed, and that this kind of statistic doesn't accomplish what it set out to do, although it's certainly interesting for its own sake. I think it's a cha
8.
by jal278 1y ago
Yeah -- I don't get why this is front-page -- reads like LLM quasi-insight: "Through activation, lifeless equations became living systems. The neuron was no longer a mere calculator; it was a decider - a locus of transformation wh
9.
by jal278 2y ago
The idealized (Science 1) / realpolitik (Science 2) dichotomy is both real and at first depressing. I also did a PhD in machine learning, and became quite disillusioned after seeing how the sausage was made, and how different the proce
10.
by jal278 2y ago
This is naive 'populism': There's no way to avoid 'allusion' writ large -- e.g. do you object to biblical references, or to references to particular experiences that only some people have (heartbreak, death of a fat
11.
Evolution and the Knightian Blindspot of Machine Learning(arxiv.org)
2 points by jal278 2y ago|0 comments
12.
by jal278 2y ago
> Despite the HN comments complaining about it being overwhelming and a dark reflection of how awful and distracting the internet is, clearly enough people enjoyed it to get to the front page. Is this like a massive HN wooosh -- how can
13.
by jal278 2y ago
> High intake of sweetened beverages was associated with higher risk for most of the studied outcomes, for which positive linear associations were found. In contrast, a low intake of treats was associated with a higher risk of all the st
14.
We Need Positive Visions for AI Grounded in Wellbeing(thegradient.pub)
1 points by jal278 2y ago|0 comments
15.
Interactive poetry breeding through Mixtral base model LLMs(flourish.ing)
2 points by jal278 2y ago|0 comments
16.
by jal278 2y ago
But it may scale -- that's science in progress
17.
Identifying Life-Changing Books with LLMs(flourish.ing)
1 points by jal278 2y ago|0 comments
18.
by jal278 3y ago
I don't get this line of logic -- of course software has safety implications, because people use it for things in the real world. It isn't "math' that is cleanly separable from the rest of humanity; its training data com
19.
by jal278 3y ago
But applied mathematics can have ethical impact -- e.g. the concept of whether a human should trust the output of a particular language model. So GP's idea of 'trust' not applying because an object has its basis in math seems
20.
by jal278 3y ago
Where is the firm dividing line that separates math from philosophy?
21.
Machine Love(arxiv.org)
1 points by jal278 3y ago|0 comments
22.
by jal278 3y ago
The long-term impact of this paper has confused me from a technical lens, although I get it from a political lens. I'm glad it brings up the risks from LLMs but makes technical/philosophical claims which seemed poorly supported an
23.
Machine Love(arxiv.org)
1 points by jal278 4y ago|0 comments
24.
by jal278 4y ago
Author here -- some fun videos from the paper: A gallery of evolved robots: https://y2u.be/QNyNtvwA9FI An evolutionary lineage from an initial (barely-functional) seed to a final robot: https://y2u.be/M9pAJu
25.
Truthers and Tweets: Analyzing Conspiracy Uptake Among Presidential Followers(worldbetter.svbtle.com)
1 points by jal278 10y ago|0 comments
26.
Who Do Twitter Followers of Presidential Candidates Follow?(worldbetter.svbtle.com)
1 points by jal278 10y ago|0 comments
27.
by jal278 10y ago
Ideally, meditation would spur you on to action; that's the direct aim of engaged Buddhism [1]. But more broadly, many Buddhist schools aim to encourage a direct feeling of love for all sentient beings, which if combined with the philo
28.
Roll your own political poll (MTurk+Python+Pandas)(worldbetter.svbtle.com)
1 points by jal278 10y ago|0 comments
29.
by jal278 10y ago
The mechanism of introducing an information bottleneck, e.g. by changing from prose to poetry and trying to recover the prose, seems similar to autoencoder techniques that are popular in machine learning.
30.
Roll Your Own Political Poll on Mechanical Turk(worldbetter.svbtle.com)
3 points by jal278 10y ago|0 comments

More ›