biased estimator

deep learning and big computers

[biased estimator @selfattentive]

I cannot overstate how crowded "ai research" has gotten and how low quality the published material is becoming as a result. Nearly every week now I am seeing throw away experiments my lab looked into 1 or 2 years ago being dressed up as "research" and posted to arxiv.

[biased estimator @selfattentive]

@DeepDishEnjoyer I think it works because training a big neural net is like doing greedy search in a space with no(few) local minima

[biased estimator @selfattentive]

@DeepDishEnjoyer The op is interaction bait and incorrect, but I don’t think the universal approximation theorem actually explains much at all about why current deep learning works well.

[biased estimator @selfattentive]

@rishabh16_ Depends on the model. If you have a dna/rna language model with a causal mask yes you can do this. With a bidirectional mask interpreting attention weights is ~impossible.

[biased estimator @selfattentive]

@macrocephalopod I have a very serious quant trading strategy™️ I promise we just have to wait for the college football season to start trading it.

[biased estimator @selfattentive]

I’m sure they’re good at hardware or whatever but every time I’ve had a conversation with one of their ml people I came away unimpressed.

[Shay Boloor @StockSavvyShay]

Cathie Wood bought ~$35M of $CBRS today

[Guy @nosilverv]

great moment in every optimizer’s life when he finally runs the EV calc on running EV calcs on everything, realizes the whole thing has been catastrophically negative EV, deletes the spreadsheet and goes outside

[ueaj @_ueaj]

"Attention is just a special case of <abstract math thing> so we generalized it by <neglecting the other 30 abstractions and conditions required for frontier architecture> and we found it performed <p hacking> compared to <naive baseline>"

[biased estimator @selfattentive]

@mandylu Interaction bait

[biased estimator @selfattentive]

@Tyler_A_Harper This is like saying universities shouldn’t have computers or microscopes or any one of the countless other tools that make research and learning possible.

[Miles Brundage @Miles_Brundage]

Bay Area people are like "become rich. go into debt if you have to"

[Andrew Gordon Wilson @andrewgwils]

When someone says "we need theory of deep learning", note that probably nothing will "count" as a theory of deep learning unless it is *their* theory, or unless it is speaking a language, and using techniques, that they already have a bias towards.

[biased estimator @selfattentive]

@Tyler_A_Harper “Why are they paying for access to the most revolutionary technology of our lifetime? That money should go to paying fat narcissists to stalk ucpd officers and paint graffiti all over Hyde park”

[biased estimator @selfattentive]

@Tyler_A_Harper Well at least Claude enterprise subscriptions will produce something of value

[biased estimator @selfattentive]

@JacquesThibs I don’t see any reason why this would be the case

[biased estimator @selfattentive]

@Lib_Development What a way to spend time.

[biased estimator @selfattentive]

@akarlin French should be bumped up and German should be bumped down

[biased estimator @selfattentive]

Milton Friedman

[Phil Hoyeck @PAHoyeck]

Since everyone is sharing their favourite American authors, who is the best American philosopher?

[biased estimator @selfattentive]

@PAHoyeck @razibkhan Milton Friedman

[biased estimator @selfattentive]

@mattapplepi @Grnfink2 Seems like it puts the trainee into desperate and violent state of mind that probably comes in handy

[biased estimator @selfattentive]

@mattapplepi @Grnfink2 I don’t know anything about military stuff but even if it isn’t a practical skill I bet this kind of training is good for teaching a mentality that is useful in war