samrus

joined 11 months ago
sorted by: new top controversial old
Bill Gates told a German newspaper that GPT5 wouldn't be much better than GPT4: "there are reasons to believe that we have reached a plateau" [N] in c/machinelearning@academy.garden
[–] samrus@alien.top 1 points 10 months ago

embeddings are statistics. they evolved from linear models of statistics but they are now non-linear models of statistics. Bengio 03 explains this

[The Athletic] Arsenal not planning to end Visit Rwanda deal despite court ruling against government’s asylum policy in c/main@soccer.forum
[–] samrus@alien.top 1 points 10 months ago (1 children)

Egypt - No more songs about Salah.

what did egypt do? legit asking

[D] What AI topics are you curious about but rarely see in the spotlight? in c/machinelearning@academy.garden
[–] samrus@alien.top 1 points 10 months ago

its somewhere in the neurons. would cost the company alot to get to it though. best not to worry about it /s

[D] What AI topics are you curious about but rarely see in the spotlight? in c/machinelearning@academy.garden
[–] samrus@alien.top 1 points 10 months ago

i hope for your sake this is a deliberate and self-aware troll

England men's squad named for Malta and North Macedonia in c/main@soccer.forum
[–] samrus@alien.top 1 points 10 months ago

what did Ben White do? pretend to support the LGBTQ community and then stab them in the back for a payout? no ... that can't be it

1
[D] model for "understanding" dashcam images (alien.top)
 

i have some dashcam footage from my car and want to see how a model could embed the images in an unsupervised (or self-supervised) way so i dont have to label everything.

like if scenarios that are semantically similar, but different in pixel space (pulling out the driveway in the day versus pulling in at night) could be clustered close-ish together in latent space so that i could label fewer images and have the model get the other using something like k-nearest or whatever.

i am starting off with just frame level before i try to tackle videos as a sequence of images (will probably lose interest by that point, so want to get images dealt with first). i looked in to VAEs and tried training one from scratch on my data but i dont have enough compute power for that.

does anyone here have any ideas about this? any pretrained off the shelf models that i could use for this? any leads for a literature survey?

[theathletic] Manchester City have dedicated a training pitch to former club captain Ilkay Gundogan as a tribute to his services for the club. in c/main@soccer.forum
[–] samrus@alien.top 1 points 11 months ago

tinpot club. arsenal built a statue for wenger, who the club is names after, just this year

Karim Benzema is considering taking legal action against the Interior Minister, who presented him as having "notorious links with the Muslim Brotherhood". in c/main@soccer.forum
[–] samrus@alien.top 1 points 11 months ago

wait till you graduate from high school. it'll be very exciting seeing all this stuff for the first time

Karim Benzema is considering taking legal action against the Interior Minister, who presented him as having "notorious links with the Muslim Brotherhood". in c/main@soccer.forum
[–] samrus@alien.top 1 points 11 months ago

do you think civil suits are you tattling to the government?