Manta Array

@tim_lees_burner @rachdele How dare you, we have activation functions

@jawwwn_ Softer-harder (cortisol) energy

Palantir CEO Alex Karp’s best longevity hacks: “Do technical Norwegian-style cardiovascular training.” “I think something meditative— I do Tai Chi every day… something that rebalances your software-hardware energy.” “My best work is done if I get 8 hours sleep.”

It's smart to be skeptical of scientific papers vetted and gate kept by editors with undeclared conflicts of interest, with advice from anonymous, unpublished peer reviews, with reviewers hand picked by the editors. Independent replication is the better way.

ICYMI: NIH must re-earn trust through structural transparency, not rhetorical reassurance and splashy journal publications. In short, our science must be replicable and the faith Americans place in our expertise to provide that science should be auditable. I spoke at length last week on how NIH can reclaim the public’s trust at a Washington D.C. roundtable discussion held by the @MAHA_Institute. ➡️ bit.ly/4aAVpif

@NIHDirector_Jay They focus on it because they want to distract from their glaring shortcomings like millions wasted on Alzeimers fraud and many other examples

The Hill is more interested in mischaracterizing my remarks than engaging with the facts. Secretary Kennedy and I are fully aligned in their determination to identify the root causes of autism and confront it as the serious public health crisis it is. NIH is doing exactly that by conducting rigorous research through the Autism Data Science Initiative, following the evidence as it develops, and refusing to prematurely rule anything in or out. We will not be distracted by media spin. We will follow the science wherever it leads, period. bit.ly/4aflm5w

@Sam_kuyp The toxic arguments that used to break out on the PBS spacetime discord server were really something

People often claim that many-worlds is adopted because its ostentatiousness is alluring. But I’ve never met anyone who adopted it for any reason other than the arguments. If anything, people seem put off by how outlandish it sounds, making it projection on their part.

@gniksAtsuJmi @Sam_kuyp No, the opposite it reconciles subjective and objective probability

@Sam_kuyp It makes the probabilities meaningless.

Every time I return to many-worlds, it seems so natural, so consistent with the rest of physics, and so perfectly attuned to the mathematics of quantum theory, measurement, and entanglement, that I find myself genuinely puzzled as to why it remains so controversial.

@mattyxb it will work bc the same roommate will also be too lazy to throw out an apparent empty box

I invented a new pizza box for people who have room mates who are always stealing their leftovers.

@KordingLab Edge of chaos neuro social bonding

Why is it real fun in a mosh pit? It is hard to see normative nor mechanistic reasons.

@NewsLambert I thought it was Blackwater mercs buying all dem houses

The fact that the public has confused Blackstone with BlackRock, and made BlackRock—which doesn't buy single-family rentals—the face of institutional homebuying, will never not be funny

@JDVance @SecKennedy that has essential amino acids and probably essential fatty acids

hey @SecKennedy this new food pyramid is solid but you forgot to include one dietary staple:

@dezgoldie tbf they teach this in defensive driving training, it is backed by data that it is safer for assets and humans/animals

People who MUST park backwards every time they park is a very interesting psychological profile.

@GeorgiaTechFB wish i could like post 3X

Good company in the office today 🦾 #StingEm 🐝

@anon_opin oh no the horror. hey google call mom, hey google what is 45 X 450

The fact that calculators and computer keyboard numeric keypads have 789 across the top, while phone number keys have 123 at the top, is one of the greatest user interface design failures of all time.

Finally! Another key paper by Sylvain Lesné about the purported key Alzheimer’s-related protein amyloid-beta *56 is retracted. Validates findings by @Thatsregrettab1 @MicrobiomDigest @schrag_matthew et al science.org/doi/full/10.11… See my original story science.org/content/articl…

A drone that flies through the sky — and swims underwater. Yes, you read that right. One student just built it. Andrei Copaci created this hybrid drone for his Bachelor’s project — 3D-printed, self-coded, and powered by variable-pitch propellers that adapt mid-flight to move seamlessly between air and water. And it actually works. To me, this isn’t just a cool gadget. It’s a wake-up call. We’re watching the line between “student projects” and “industry breakthroughs” disappear. One person, a laptop, and a printer can now do what once required millions in funding and government labs. So here’s the real question: If innovation is becoming this accessible — what happens to companies and industries that still think disruption comes only from big budgets? #AI #Innovation #Technology #Robotics #FutureOfWork #STEM

I do like this paper quite a bit -- having a bunch of different methods be special cases of the same generalized information bottleneck framework is cool.

@ChrisWLynn @matthew_pines Barandes has not shown what QM really is as this post claims, Barandes would not even agree with this

@matthew_pines Barandes’s work is very interesting but the vast majority of physicists still think quantum mechanics is Markovian. Either way, we should understand these non-Markovian dynamics better!

Barandes (Harvard) would disagree that physics is Markovian, since he’s shown that quantum mechanics is really just a stochastic process whose “indivisible” rules can’t be broken into independent time-steps. This is a non-Markovian generalization of textbook stochastic processes.

@matthew_pines He has not shown that, he has a developing idea and work in progress with many holes to fill

@ChrisWLynn @MattLeighton5 thanks, look forward to reading. i worked on similar topic and i just think it is important to distinguish ontological deviations from Markovian dynamics and epistemic deviations. i.e. observation layer of even a simple HMM is non-Markovian

@ArrayManta @MattLeighton5 By make sense I just mean how can we understand these complicated dynamics? Almost all our intuitions are based on Markovian dynamics

Physics is Markovian: the next state only depends on the previous state. Biology is fundamentally different: each state can depend on details far into the past. How can we make sense of these non-Markovian dynamics? Two new preprints from @MattLeighton5 work toward an answer