> On October 23, 2001, U.S. representative Jim Sensenbrenner (R-WI) introduced House bill H.R. 3162, which incorporated provisions from a previously-sponsored House bill, and a Senate bill introduced earlier in the month.[5] The next day, October 24, the Act passed the House by a vote of 357–66,[6] with Democrats comprising the overwhelming majority of "no"-votes.

> The bill, which removed most of the changes from the Senate version, passed Congress on March 2, 2006, and was signed by President Bush on March 9 and 10 of that year.

Worth noting a different manifold "exists" after each transformation (e.g. layer). You only sample from the same manifold when you apply the same transformation(s).

TL;DR Picard's initial arguments are pretty weak, even admitting that Riker as opposing counsel almost had him convinced. During a recess Picard talks to Guinan where she alludes to the future subjugation of many Datas which Picard connects to slavery. Back in the courtroom Picard calls Maddox as a hostile witness and gets him to define sentience--intelligence, self-awareness, consciousness--then walks him into conceding Data meets the first two. Picard's closing boils down to, "we don't know if he meets the third--you can call Data a toaster and rule he is property--_but what if you're wrong_". The judge rules on the basis of erroring on the side of caution due to that uncertainty. It's really a great scene.

We're not there yet, obviously. No LLM brings Data's level of awareness but it's as relevant a story as ever because it isn't really about AI but othering for the purpose of subjugation.

[0] http://www.chakoteya.net/NextGen/135.htm

Apparently a combination of Mie and Rayleigh scattering.

- https://en.wikipedia.org/wiki/Mie_scattering

- https://en.wikipedia.org/wiki/Rayleigh_scattering

They specify early to mid career. Imo they're anticipating a ton of applications and bounding it makes reviewing them tractable.

Is the pride not in solving the users' problems?

> nobody talks about it, treats it with interest, or pays above market rate to work on it.

Definitely needs a citation for this one. For so many products the user isn't paying for standout design. They're paying for insight, leverage, velocity, convenience, whatever. The market definitely supports this by paying above market salaries.

Good design can be a useful differentiator but it isn't the only way for a tool or product to "spark joy" and often _fancy_ design (not good design) is used as a crutch for a subpar product.

Then, if you're like me and read this years ago, play around with the Light Mode dropdown which was new to me. :)

EMA helps because it changes more slowly than the learning network which prevents rapid collapse by forcing the predictions to align to what a historical average would predict. This is a harder and more informative task because the model can't trivially output one value and have it match the EMA target so the model learns more useful representations.

EMA has a long history in deep learning (many GANs use it, TD-learning like DQN, many JEPA papers, etc.) so authors often omit defense of it due to over-familiarity or sometimes cargo culting. :)

The online, continuous and lossy version of this problem is more like how our memory works and still largely unsolved.

[0] https://arxiv.org/abs/2112.04035

[1] https://arxiv.org/abs/1803.07770

[2] https://www.nature.com/articles/s41586-018-0102-6