owenshen24

5 ms·

1.
by owenshen24 5d ago
> For an agent TAU is the score that matters and a 20 point swing is not noise. +1, stuff like this is a big turnoff for reading
2.
Another cyberattack by internal OpenAI agents targetting RubyGems(twitter.com)
4 points by owenshen24 5d ago|0 comments
3.
by owenshen24 12d ago
in general, the largest consumers of ai services seem to ask for more capabilities. i wish there was more demand for safety from users. i also wish that these types of illicit system usage would be met with punitive action the same way a hu
4.
by owenshen24 12d ago
https://news.ycombinator.com/item?id=49565071 > OpenAI officials learned of the incident weeks ago but kept it under wraps as executives grappled with the fallout from the July breach of the open source repository Huggin
5.
Donanemab in Early Symptomatic Alzheimer Disease(jamanetwork.com)
1 points by owenshen24 3y ago|0 comments
6.
Opinions on Interpretable Machine Learning and 70 Summaries of Recent Papers(alignmentforum.org)
2 points by owenshen24 5y ago|0 comments
7.
NoCode MBA Program(nocode.mba)
1 points by owenshen24 6y ago|0 comments
8.
by owenshen24 6y ago
Is there good reason to think that optimizing resumes will help with the job searching process? I've heard that many resume screens are automated these days, which has led to my hearing stories of people trying to game this by cramming
9.
Interpretability in Machine Learning: An Overview(thegradient.pub)
1 points by owenshen24 6y ago|0 comments
10.
by owenshen24 6y ago
Agreed. The latency between keypress and result makes it such a pain to use.
11.
by owenshen24 6y ago
Agreed. After writing my own static site generator in Python, I went back and was impressed with all the options available in Pelican. Lots of things I wished I had built were included.
12.
by owenshen24 6y ago
To add some more context, here's a rather readable summary: https://www.greaterwrong.com/posts/FRv7ryoqtvSuqBxuT/underst...
13.
by owenshen24 6y ago
SSC has an interesting take on theodicy that addresses this claim: https://slatestarcodex.com/2015/03/15/answer-to-job/
14.
by owenshen24 6y ago
Doesn't seem like a well-curated list. Some of these, e.g. Uber's God Mode, don't even fall under the normal usage of the term "AI". Others, like Palantir or WeChat, seem to denigrate an entire company (rather than
15.
by owenshen24 6y ago
Trees aren't enough to scale up. You can play with toggling deforestation and reforestation here: https://en-roads.climateinteractive.org/scenario.html?p50=0.... Even in the most optimistic scenarios, it's clear t
16.
MicroCOVID Project, Micromorts but for Covid(microcovid.org)
16 points by owenshen24 6y ago|2 comments
17.
by owenshen24 6y ago
I wrote about something similar for self-help, more broadly here about how good self help is self-defeating: https://mlu.red/55070266310.html
18.
by owenshen24 6y ago
Dustin is also notable for starting Good Ventures ( https://www.goodventures.org/about-us ) which funds a lot of research and philanthropy into some great causes.
19.
by owenshen24 6y ago
I see. Thanks for adding to the clarification. I think that the presentation of nulls as "results" can definitely be disingenuous. Ideally, science would have a better database to keep track of what people find, where we could add
20.
by owenshen24 6y ago
I definitely think there's more room for this sort of guided / ML analysis, but I'm not quite sure to make traction on extracting the structure of scientific papers...hopefully someone with more experience can chime in. I thi
21.
by owenshen24 6y ago
Oof, that sucks to hear. Thanks for letting us know it's not been updated.
22.
by owenshen24 6y ago
That sucks to hear. Null results are important, as you say, if only to dissuade others from doing the same. See also the "file-drawer problem" ( https://en.wikipedia.org/wiki/Publication_bias ). Also, with rega
23.
by owenshen24 6y ago
Yeah, my bad. I mis-remembered. Thanks for clarifying!
24.
by owenshen24 6y ago
Fogg has some good thoughts on positive behavior change. James Clear, who wrote Tiny Habits, is another good person in this space who writes reasonable stuff. Self-plug for an overview on research into the habit formation literature that I
25.
by owenshen24 6y ago
I haven't read Giridharadas's book, and this article was very light on substance. I do think this is an interesting topic. Effective altruism has lots of ongoing internal discussion about the efficacy of pushing for systemic chang
26.
Habits: Theory, Creation, and Removal(mlu.red)
2 points by owenshen24 6y ago|0 comments
27.
by owenshen24 6y ago
The notion of "Dead Child Currency" ( https://www.lesswrong.com/posts/6cRhG6PKeASdNHqxD/dead-child... ) may intrigue you.
28.
Interpretability in Machine Learning: A Broad Overview(mlu.red)
1 points by owenshen24 6y ago|0 comments
29.
by owenshen24 6y ago
With the rise of highly effective black box models, there's been a rise in interest towards interpretability for machine learning. This is an area I'm excited about, so I did a dive into the existing research. I found a few surpri
30.
Interpretability in ML: A Broad Overview(mlu.red)
1 points by owenshen24 6y ago|1 comments

More ›