kriro

34 ms·

331.
by kriro 6y ago
Very interesting comment, thanks for taking the time to write it :) I think if memory is the only problem than optimizing training time should be more of a concern. I'm imagining a huge language model than can retrain very quickly. So
332.
by kriro 6y ago
I'm looking down at my half full glass and all I can think is....fantastic news. Remote work means I can take any job I want.
333.
by kriro 6y ago
I also started with QBasic (with no internet connection). Built a small choose your own adventure. Then I wanted to build a "password program" that would ask for a password before you could get "into the computer". I hea
334.
by kriro 6y ago
That's a bit too idealistic. Maybe some of the stuff was started this way but by and large all hobby projects that are now widely used depend on paid people working on it full time/a lot. The Linux kernel is a great example. But I
335.
by kriro 6y ago
I haven't looked into UE in quite some time as I use Unity for most of my stuff (not a game developer, just building quick prototypes for AI, simulations, dabbling in AR etc.) but the demo looks pretty stunning. I tend to do a technolo
336.
by kriro 7y ago
Maybe it's all subjective but I am willing to pay a premium for Kerrygold (Irish) because I read once that the cows are grass fed and am imagining the butter tastes a lot better. It's very interesting how much the price can fluctu
337.
by kriro 7y ago
"""You see it in manufacturing. Contrary to conventional wisdom, American manufacturing output is higher than ever, but why has so much manufacturing been offshored to places with cheaper manual labor?""" &quot
338.
by kriro 7y ago
Being an uncle of 10 month now I have made a list of the things to keep in mind when I decide to have my own children. It's pretty short: - Spend as much time with your child as possible and let them roam free. Just have a loose eye ou
339.
by kriro 7y ago
The FED announced a huge new stimulus package (as did the Bank of England earlier today). I'm also surprised because I'd expect stocks to tank because that's what would make economic sense to me. People at home, not spending
340.
by kriro 7y ago
"""Risk quantification attempts to assign numeric values to risks, instead of qualitative labels such as "Critical" and "High".""" Nitpicky...but... Shouldn't risk always be quantified?
341.
by kriro 7y ago
My mother just finished sowing 100 washable ones so they can use them in the elderly care home she works in (she used the guide provided by the city of Essen, Germany). They still have some left but estimated they'll run out soon and a
342.
by kriro 7y ago
It's all linked on the website: https://tiny.cc/maskswork
343.
by kriro 7y ago
Why not sell to non-enterprise as well? Pretty sure some people would like to filter adds from their TV and replace them with something else. Maybe come up with some curated content, possibly subscription based. A good entry might be parent
344.
by kriro 7y ago
The standard argument is that a mask may give you a false sense of safety which results in subtle behavioral changes that are a net-negative. Additionally if the mask inhibits breathing you might take deeper breaths which is also a net-nega
345.
by kriro 7y ago
I think they have handled it pretty well. I'm pretty sure they spent some time talking to experts. I've worked in a research project on crisis management before and have a very high opinion of the Robert Koch Institute which was o
346.
by kriro 7y ago
I think the first question that you will face is if you agree with the Efficient Market Hypothesis (EMH) and the random walk. Basically they say that prices reflect all available information and that you cannot "beat" the market (
347.
by kriro 7y ago
I've sold some AR projects to customers. Just product visualizations for trade fairs etc. nothing too fancy. It's cool and fun to work with but my takeaway was that for many practical applications it's pretty cost/time i
348.
by kriro 7y ago
That seems like a pretty unrealistic scenario. I guess it's a thought experiment but positioning (and the existence of branding/marketing) alone would indicate to me that Barry and Mark are not going to be exactly the same. Positi
349.
by kriro 7y ago
I'm using DDG but I have to "!g" a lot. The English language results are quite good but the German results are often not what I'm looking for. I'm assuming this will improve with time so I'm not too worried but
350.
by kriro 7y ago
As an interesting linguistic side note in the German translation the Elves are called Elben. I believe this term is only used for Tolkien-Elves. The term Elf/Elfen exists in German as well and is commonly used in fantasy/RPGs to d
351.
by kriro 7y ago
Sometimes I feel like I have a "genetic advantage" to not be attracted by social media. I have a Twitter account that I rarely use (mostly for API-access for NLP research), I check into my Facebook account maybe once a month and h
352.
by kriro 7y ago
I guess it depends a bit which field they work in exactly. I'd be rather surprised if rigorous DL researchers in NLP haven't heard of him because I expect "Universal language model fine-tuning for text classification" (a
353.
by kriro 7y ago
I used to work in a research project where I went to some UX conferences in Germany (not really my field). The big irony/running gag is that SAP seems pretty good in UX-research, Hasso Plattner Institute is fairly recognized for their
354.
by kriro 7y ago
My first lesson in the real world after I graduated was that the big name consulting companies are in many instances just well managed brand names. They are often used by management to "objectify" decisions tat are already made. S
355.
by kriro 7y ago
As a counterpoint I'd argue that these movies are usually also successful in countries where they are screened in non-English. Those voice actors are usually unknown to the general audience (one could argue that they are often the asso
356.
by kriro 7y ago
I'm going to try this simply because spaCy is so great. It's my goto for NLP in different languages (notably German) and a joy to work with.
357.
by kriro 7y ago
I'm curious. How do people protocol their experiments? When I started, I used to just keep the cells but that lead to very long and impossible to parse Jupyter notebooks. I have since opted for keeping a journal.txt file in Atom where
358.
by kriro 7y ago
Direct link to the preprint: http://web.eecs.utk.edu/~azh/pubs/Chattopadhyay2020CHI_Noteb... Interesting study, I like the mixed-method approach. A quick glance at the industry of the participants suggests that th
359.
by kriro 7y ago
I would reserve the question of ethical A.I. until we are approaching strong A.I. As of now, where A.I. (specifically ML/DL) is mostly used to solve isolated problems the question of ethics is more about the use of the technology. Tech
360.
by kriro 7y ago
I really enjoyed Fravia's page back in the day and one of my all time favorite sites by +Malattia (had to search a bit to find it): http://3564020356.org/ Which reminds me...I never figured out that hash-maze but back

More ›