NeurIPS 2020 “English to Bash” Competition Goes Live | Hacker News Reader

NeurIPS 2020 “English to Bash” Competition Goes Live

nlc2cmd.us-east.mybluemix.net

87 pointsArtWomb17 comments

"Try it out" live demo loads forever for me. Possibly overwhelmed by requests?

_edit_

I wonder how they handle different distros having different versions of tools (say netcat has at least 3 popular versions with different flags).

The competition is specifically constrained to a specific version of ubuntu and provides the following reference:

http://manpages.ubuntu.com/manpages/bionic/en/man1/

> "Try it out" live demo loads forever for me. Possibly overwhelmed by requests?

If you're using Firefox, try to disable Enhanced Tracking Protection. It did the trick for me.

Ohh, thank you! Didn't realize that was the case.

Of course everyone will try "recursively remove root"

No no no no. You don't remove root. You remove ../../../../../../../.. .

Or "remove eight directories up from the current directory", I guess.

It'd be impressive enough to recursively traverse a tree using memoization in bash.

It doesn't let you select the result to ctrl+c.

Just try GPT-3...

So dubious bash completion requiring a cluster of GPUs in a a closed cloud and the energy of a hydroelectric dam to run?

Does it really matter how it’s solved the first time?

The second version can require three orders of magnitude less hardware. Premature optimization is the root of all evil.

Isn't the whole point of GPT-3 to have tons of data and hope for the best by training it with a huge amount of time and energy ?

For training -- using the trained algorithm is cheap. It's a one-time cost (well, one-time in the final version)

You do realize that reducing costs means dropping parameters. So what you are suggestion is reversing the very technique that enables this result. Color me doubtful.

It doesn't. There exist far more sophisticated techniques to reduce the size of a neural network while minimizing loss of accuracy.

The energy required to use GPT-3 is a small fraction of the energy required to train it.

Indeed, the demos of GPT-3 should specifically tackle this.

Reply on news.ycombinator.com