Nalta

It's nuts to me that I can't tell if this is an AI response or not. I feel like 80% sure this is a robot.

that sounds like a great idea! I'll return 3 when peers ask me for a random number, and then I'll start seeing if that seed shows us in any prominent RSA private keys

fwiw, I take sites like this and use chrome's option to make them stand-alone apps. (three dots -> more tools -> create shortcut -> open as window)

in reality, more often than you'd like, the proceeds of these projects go straight to scammers who don't actually have the rights to the source itself.

For anyone interested, my whole PhD was in biomedical hypothesis generation! I think the most "serious" attempts at building these systems have been focused around providing assistance to scientists, and not just coming up with new ideas on their own.

here's an actual medical paper that my first system, Moliere, was able to help discover:

https://link.springer.com/article/10.1007/s11481-019-09885-8

My dissertation used a lot of dask code, and I found it to be an amazingly powerful tool for helping me to glue a few dozen other libraries together and orchestrate them as a single distributed system. sure it's half done spark, but if it were fully done spark I couldn't have used it as flexibly.

I feel like this is the same argument as "kids these days don't know anyone's phone number!" If it works it works

as a warning, I've heard that the big spreadsheet makers considered this and eventually passed on the idea. I believe the logic is that spreadsheet users don't understand ML, and if you want to do ML, you probably know how to export a spreadsheet.

As someone else who works on systems like these, I agree training data is the whole problem. However you can use some techniques like homomorphic encryption and gradient pooling to collect training data from client code while remaining end-to-end encryption. It's hard, but it's not impossible.

so like... what is it?

I'm curious if there's a market for people who want to buy small projects like this. I imagine there's a lot of projects that are already on GitHub with a small following with owners that would like to cash out, but I'm not sure there's a market of buyers looking to pick up repos?

hey, keep that shit on 4chan.

I think I agree with you, but then what about something like Hearthstone? How much glitz would they have to add before a digital version of the same phenomenon be considered gambling, if presentation were the metric?

(Citation Needed)

Cake recipe. Ingredients: one nice container, one serving utensil, one cake.

I really hope those grad students that she mentioned didn't get a couple years added on to their degrees as a result of this. I mean good on her for finding the error, but I can't imagine what it would be like to be told "cancel your thesis, I made a mistake and your work is now invalid."

I found this platform really easy to play around with when building a game. I put together a version of Hanabi, and the system was flexible enough to let me make player hands that are visible to everyone _except_ the player. Reconstructing this relatively simple game from scratch only took ~3h, and that is everything from finding royalty-free assets to writing the manual.

How is this different than dask?

Great! I can't read anything!

I use protos as configs pretty frequently. They have comments, easier syntax, there's a schema file the describes exactly whats expected. You can parse them from text, csv, or even awful binary, and use them in practically any language.

You could write this exact article but for the research peer-review process...

Actually, can somebody write that? :P

I would go a step further. Please don't show me code. Show me runtimes, interesting problems, and MAYBE a line that shows "Aw gee! Thats easy!"

Clearly one LSTM didn't work. Lets try FOUR!

- edit: miscounted number of LSTMs

True, but I think this could really be useful for cases similar to google flags (https://github.com/gflags/gflags) wherein each file in a library adds to the global configuration. Then your main can more easy parse out these options to be shared across the entire codebase.

I really really do not want the ability to write bugs in my bank account.

As a Ph.D. student, I found this to be an interesting outlook, albeit a bleak look at the current job market. My research is in machine learning, and I'm hoping my job search will be easier :P