domenicrosati

5 ms·

1.
by domenicrosati 4y ago
Simulation? Like fluid dynamics. I heard that was CPU intensive.
2.
by domenicrosati 4y ago
Well done
3.
by domenicrosati 4y ago
Ya... those are not great places to start with him... as far as philosophy of science essays in conjectures and criticisms and objective knowledge are really great and much more mature than the logic for the reasons you mentioned. Plus he d
4.
by domenicrosati 4y ago
Same
5.
by domenicrosati 4y ago
Well said. Attribution is important but it's impossible to know all the literature that is out there and even when you've tried your best to seek out prior arts there is always a possibility that you've missed something or no
6.
by domenicrosati 4y ago
I wonder what software would be good for a logbook like this... I just use google docs for these kinds of things. Sure wandb and jupyter notebooks are good but they are not so good for notes and ideas and documentation
7.
by domenicrosati 4y ago
I am all for independent discovery but the author should know in the language modeling community we have known this for years [1]. The authors of DALLE are well aware of it. We have many concerted efforts to try to solve for it. Heck even h
8.
by domenicrosati 4y ago
Hey there - OP here. I haven't used spyder I'll have it check it out. The primary use case is: I am a researcher in nlp where speed of prototyping is key. I work in an environment where research fragments are primarily jupyter not
9.
by domenicrosati 4y ago
Woah this is exactly what I'm looking for! Thanks
10.
by domenicrosati 4y ago
In 10+ years of startup experience on the science and engineering side. Sales fixes everything and is the hardest thing. If sales can execute then there is no problem. It seems obvious but it's really not and people make the business a
11.
by domenicrosati 4y ago
The T0 series by big science is essentially an instruct model (though using multitask prompting instead of user feedback). You should check it out. I have got very competitive results on prompting t0-11b v instructgpt3(text davinci 2)
12.
by domenicrosati 5y ago
Cool! What motivated you to write it? I find the d2l implementation nice and simple as well https://d2l.ai/chapter_attention-mechanisms/transformer.html The one in pytorch itself isnt so bad either. Be nice to see some
13.
by domenicrosati 5y ago
Be awesome if it were more comprehensive. I typed in d-serine, bromantane, nsi189, modafinil, phenibut with no results...
14.
by domenicrosati 5y ago
For folks who are interested. This is actually an already well studied topic! It is called "language complexity". The consensus is that in aggregate (all language artifacts like writing, conjugation, syntax) no language is more co
15.
by domenicrosati 5y ago
Is anyone able to find evaluation of these models besides speed and power consumption? I really cant find any baseline comparisons of accuracy and so on for any relevant datasets being used these days. I wish they did a better job of presen
16.
by domenicrosati 5y ago
Just anecdotally, Ive had a paper rejected once from arxiv. I am glad they did... it was not a very good one. Every other time thought has been fine
17.
by domenicrosati 5y ago
For brand name journals and conferences, yes they do. My experience with those is a very high rejection rate. There is a good reason to reject if you are not using a standard dataset. How can you compare the results of two approaches, to sa
18.
by domenicrosati 5y ago
Be cool to see this for all the pokemon
19.
by domenicrosati 5y ago
I think the author is not doing a good job of presenting the research program behind ever larger language models. Many researchers are interested in the effects of scale on natural language understanding tasks and generalization. One of the
20.
by domenicrosati 5y ago
https://scite.ai does this as well (also citations are classified and analyzed whether they provide supporting or contrasting arguments to the citations) The scite extension also works with connected papers so you can see that i
21.
by domenicrosati 6y ago
Done https://github.com/scitedotai/scite-zotero-plugin
22.
Scite Extension Inserts Smart Citation Badges on Academic Literature Databases(medium.com)
1 points by domenicrosati 6y ago|0 comments
23.
Show HN: Scite Advanced Citation Search(medium.com)
2 points by domenicrosati 6y ago|0 comments
24.
Scite Advanced Search: Discover Research in Context(medium.com)
2 points by domenicrosati 6y ago|0 comments
25.
The Microbiology Society and Scite Sign Indexing Agreement to Improve Science(medium.com)
1 points by domenicrosati 6y ago|0 comments
26.
Show HN: Scite Reference Check: check the reliability of your references(medium.com)
1 points by domenicrosati 6y ago|0 comments
27.
by domenicrosati 6y ago
Html
28.
by domenicrosati 6y ago
Why not use more managed services like a managed DB and some PaaS for the backend services and workers? I imagine that would be both cheaper and less overhead (in ansible yaml) ect.
29.
Integration Testing with Golang(harrigan.xyz)
1 points by domenicrosati 7y ago|0 comments
30.
Manifold: Marketplace as a Service(techcrunch.com)
3 points by domenicrosati 7y ago|0 comments

More ›