sockfish

4 ms·

1.
by sockfish 4y ago
I've always wondered if machine learning would help here - we already have language agnostic embeddings (LaBSE, LASER) - now all we need is a way to pronounce them. Perhaps you could learn one by training an autoencoder that is constra
2.
by sockfish 5y ago
Yeah but one of the BSL2 labs (coronavirus research was being done in BSL2/3 labs, not BSL4) is not. It is a couple of hundred meters down the street https://www.google.com/maps/d/u/0/viewer?mid=1D1
3.
by sockfish 5y ago
It does also explain why they are not found all over the Roman empire but mostly in the north (the article mentions England, Germany, and the Netherlands) where it is probably COLD for the average Roman legionnaire
4.
by sockfish 5y ago
I often wonder why so much effort is being put into shoehorning everything into a single language. Wouldn't it make much more sense to use a fully differentiable DSL for machine learning / xla, then call it from whatever host lang
5.
by sockfish 5y ago
This experiment has been done in Rotterdam, the Netherlands in the 90s. It gained a tonne of publicity but it did not end up getting widely adopted. Most homeless people (understandably) feared getting stigmatized if they were spotted sleep
6.
by sockfish 5y ago
I wonder if you can play chess in your mind without being able to visualize the board.
7.
by sockfish 5y ago
"By that definition the USD is a ponzi. No, national currencies too fail to fit the definition, because people do not "invest" in them with the expectation of gain" <-- The 6.6trillion/day forex market begs to di
8.
by sockfish 5y ago
Roughly O(log n) by the looks of it.
9.
by sockfish 5y ago
I struggled with the accounting concepts until I stumbled on this blogpost from Martin Kleppmann: https://martin.kleppmann.com/2011/03/07/accounting-for-compu... . After that I finally 'got' the acco