|
https://twitter.com/oh_that_hat/status/1349717629838221316
|
#
?
Jan 15, 2021 03:13
|
|
|
|
| # ? Apr 27, 2024 19:34 |
|
|
https://twitter.com/sam_power_825/status/1375434434862845954
|
|
#
?
Mar 26, 2021 15:56
|
|
|
|
|
#
?
Mar 26, 2021 17:04
|
|
|
so is machine learning anywhere close to self-driving cars or old muskie’s press hype or is it still a scam?
|
|
#
?
Mar 31, 2021 03:25
|
|
|
right now johnny cabs are as viable as musk's rockets are reusable
|
|
#
?
Mar 31, 2021 03:48
|
|
|
true “self-driving cars” require solving artificial general intelligence, they’re not something you can develop algorithmically and machine learning isn’t a scam, it just isn’t what it’s sold to lay people as: it’s black-box pattern recognition systems implemented via neural networks, not “intelligence” of any sort like you can train a neural net to play Super Mario Bros. but it won’t be able to play a level it’s never seen before without a whole ton of extra work that could be to increase the size of the network by several orders of magnitude and doing tons of randomized training for both positive and negative feedback so it’s more likely to not just randomly flake out on a new situation or it could be by making the recondition models just one aspect of a system that’s under control of some kind of “executive” that can make higher level goal-oriented decisions while integrating new environmental information either way the important part is that even most recognition models are temperamental and we don’t always get why they work, what cues they’re going off and so on, and the state of the art in implementing “executive function” requires an enormous amount of compute power and isn’t very usable for arbitrary tasks yet either
|
|
#
?
Mar 31, 2021 05:54
|
|
|
self driving cars are 5-10 years away, just like they've been my entire life
|
|
#
?
Mar 31, 2021 06:09
|
|
|
this is the current state of the art in image recognition 
|
|
#
?
Mar 31, 2021 06:12
|
|
|
everybody's gonna be mocking lovely Tensorflow reinforcement learning rigs right up until they get exploded by "malfunctioning" military stealth drones
|
|
#
?
Mar 31, 2021 07:09
|
|
|
https://twitter.com/lux_forest/status/1376312075774013446?s=21
|
|
#
?
Mar 31, 2021 08:16
|
|
|
animist posted:everybody's gonna be mocking lovely Tensorflow reinforcement learning rigs right up until they get exploded by "malfunctioning" military stealth drones malfunction is that target recognition has been outsourced to child labor in some developing country, kids got bored and started randomly clicking poo poo because hey nobody checks
|
|
#
?
Mar 31, 2021 09:45
|
|
I CANNOT EJACULATE WITHOUT SEEING NATIVE AMERICANS BRUTALISED!
|
isn't that the plot of ghost in the shell 2
|
|
#
?
Mar 31, 2021 10:13
|
|
|
suck my woke dick posted:malfunction is that target recognition has been outsourced to child labor in some developing country, kids got bored and started randomly clicking poo poo because hey nobody checks actually if someone clicks outside of what the algorithm expects, that's an error and it's discarded all the "training data" is just model reinforcement
|
|
#
?
Mar 31, 2021 14:48
|
|
|
*points at logistic regression* machine learning
|
|
#
?
Mar 31, 2021 14:57
|
|
|
Carthag Tuek posted:actually if someone clicks outside of what the algorithm expects, that's an error and it's discarded Amazing(ly dumb)
|
|
#
?
Mar 31, 2021 15:00
|
|
|
morally obligated to lie on this one
|
|
#
?
Apr 1, 2021 01:59
|
|
|
redleader posted:morally obligated to lie on this one
|
|
#
?
Apr 2, 2021 02:58
|
|
|
https://twitter.com/__apf__/status/1377614677832699906
|
|
#
?
Apr 2, 2021 03:50
|
|
|
|
| # ? Apr 27, 2024 19:34 |
|
|
https://twitter.com/vamchale/status/1378834359625601024
|
|
#
?
Apr 4, 2021 23:27
|
|