jonas_kgomo

9 ms·

61.
Ask HN: Why are all companies using hiring websites
3 points by jonas_kgomo 4y ago|1 comments
62.
Minsky's Negative Expertise(twitter.com)
4 points by jonas_kgomo 4y ago|0 comments
63.
Algorithmic Efficiency Initiative(entropic.vercel.pub)
1 points by jonas_kgomo 4y ago|1 comments
64.
by jonas_kgomo 4y ago
Call for algorithmic energy efficiency initiative: it’s estimated that GPT-3 consumed 936 MWh, which would be sufficient to power ~100000 European households for a day. Ethereum has 99% efficiency by changing their algorithm. Is there a way
65.
by jonas_kgomo 4y ago
Beyond racism some of these things happen a lot in startups. As a black person as well, i might have experienced this in a much lower stakes, for example, hired at a startup. https://news.ycombinator.com/item?id=22091777 Th
66.
Chaos Machine Learning(jonaskg.substack.com)
5 points by jonas_kgomo 4y ago|0 comments
67.
Ask HN: What part of your product would you like as a white-label solution?
1 points by jonas_kgomo 4y ago|0 comments
68.
Ask HN: How would you approach an Intelligence Agency to join them
4 points by jonas_kgomo 4y ago|10 comments
69.
by jonas_kgomo 4y ago
Sampling in hip hop is about taking 1 or 2 audio sources mapping them into some new audio, SD was trained on 2.3 billion data points. In a diffusion model you put noise on an input image, until its recognizeable. For a song this would mean,
70.
A thread on Simulacrum and AI content generation(twitter.com)
1 points by jonas_kgomo 4y ago|0 comments
71.
Ask HN: How to Deal with Being Unlikeable
29 points by jonas_kgomo 4y ago|14 comments
72.
Are we in a Stable Diffusion Renaissance(twitter.com)
1 points by jonas_kgomo 4y ago|0 comments
73.
by jonas_kgomo 4y ago
I was at Palintir, and one of the things I learned from visiting is that they thrive in times of war, pandemics and recessions. Why aren't companies designed to be antifragile like this
74.
by jonas_kgomo 4y ago
I was thinking the same, a generalist looks like an even way to divest ones talents. It's a good perspective that often is talked about in effective altruism, through 80,000 hours, the question is, how should one spend their career tim
75.
by jonas_kgomo 4y ago
That is indeed my question. I want to remain a generalist, mainly due to the fact that its a big part of my character and specialising often doesn't consider whether the technology will be viable in a few years. Generalising means that
76.
Ask HN: Is There a Place for Generalists
42 points by jonas_kgomo 4y ago|27 comments
77.
by jonas_kgomo 4y ago
haven't seen anything yet, my email is not the same as username :)
78.
by jonas_kgomo 4y ago
This is super interesting, I think this is brilliant way to visualise the ML space, ofcourse there are some topics like topological data science that are not mentioned there like scikit-TDA. But so far its fascinating. AI Bots can help gen
79.
by jonas_kgomo 4y ago
i would like to know more about this. Is it open source? I would like to talk more!
80.
by jonas_kgomo 4y ago
I would like to collaborate on a visual collaboration platform for algorithm solving. email: jonaskgmoo@gmail.com
81.
DALL-E 2 open source implementation(github.com)
520 points by jonas_kgomo 4y ago|152 comments
82.
Ask HN: Distributed blind review site for papers
2 points by jonas_kgomo 4y ago|0 comments
83.
by jonas_kgomo 4y ago
If you are interested in Elixir, supabase is hiring and they are comfortable with unconventional hires, meaning your hobbyist approach could be a plus for them https://github.com/supabase/realtime
84.
Ask HN: Can we use tensors as a database data structure
2 points by jonas_kgomo 5y ago|1 comments
85.
by jonas_kgomo 5y ago
I would argue that computational complexity helps you branch out on many problems and fields. In terms of thinking models I would say basic understanding of theories like Moore's Law, Pareto Principle, Uncertainty principle and Matri
86.
by jonas_kgomo 5y ago
you need to login into Vimeo, then you can watch
87.
by jonas_kgomo 5y ago
I wonder, how can i make a good application to Media Lab? Would appreciate any insight, I am looking to apply under Hyperinstruments by Tom Machover
88.
by jonas_kgomo 5y ago
I guess, since i never wrote an academic paper, i do not know if it will be up to par with the methodologies, I thought every research kind of needs an advisor to offer direction, otherwise you can get rejected without knowing the reason.
89.
Ask HN: How can you write a paper without an advisor?
1 points by jonas_kgomo 5y ago|2 comments
90.
Some unresolved questions about AI
1 points by jonas_kgomo 5y ago|0 comments

More ›