Jamison Polackwich (@rjpolackwich) 's Twitter Profile
Jamison Polackwich

@rjpolackwich

Machine learning, I like math and physics, and science in general. Modular synthesizer enthusiast

ID: 1881035263

calendar_today18-09-2013 21:55:10

390 Tweet

87 Takipçi

649 Takip Edilen

mylar melodies (@mylarmelodies) 's Twitter Profile Photo

Another great Superbooth interview (kind of just a monologue) with/by Dave Smith on the history of his work from...two weeks ago: youtube.com/watch?v=qYGhzR…

Leonid Volkov (@leonidvolkov) 's Twitter Profile Photo

Putin's next (and final) battle. (This thread went viral in Russian so I decided to publish an English version as well) A long thread about what, in my opinion, is Putin's strategy right now, and the last gamble he is making to break Ukraine's resistance. 1/26

Florian Ederer (@florianederer) 's Twitter Profile Photo

“I am not a donkey, I don’t have a field.” So scoffed Max Weber when a faculty non-entity criticized him for writing outside his narrow discipline.

“I am not a donkey, I don’t have a field.”

So scoffed Max Weber when a faculty non-entity criticized him for writing outside his narrow discipline.
Will Kinney (@wkcosmo) 's Twitter Profile Photo

Of course that's your contention. You're a first year grad student. You just got finished reading some pop science author, Michio Kaku probably, you’re gonna be convinced of that until next month when you get to Sean Carroll, then you’re gonna be talking about how anybody taking

Of course that's your contention. You're a first year grad student. You just got finished reading some pop science author, Michio Kaku probably, you’re gonna be convinced of that until next month when you get to Sean Carroll, then you’re gonna be talking about how anybody taking
yonatan (@zozuar) 's Twitter Profile Photo

float i,e,R,s;vec3 q,p,d=vec3((.5*r-FC.xy)/r.y,.5);for(q=--q*rotate3D(PI,hsv(t/16.,.4,1.));i++<2e2;i>170.?e+=.001,d-=d+1.:d){o+=log(e*R+1.)/2e2;p=q+=d*e*R*.4;p=vec3(log(R=length(p))-t/8.*PI,e=-p.z/R,atan(p.x,p.y));for(s=1.;s<6e2;s+=s)e+=exp(cos(PI*(dot(cos(p*s),p/p)))-1.5)/s;}

Arvin Ash (@arvin_ash) 's Twitter Profile Photo

This is a page from a dictionary from the year 1900. That year, among the NEW words added were, "Automobile," "Electron," "Garage," "Ski," "Taxi," and "Wireless."

This is a page from a dictionary from the year 1900. That year, among the NEW words added were, "Automobile," "Electron," "Garage," "Ski," "Taxi," and "Wireless."
Martin Bauer (@martinmbauer) 's Twitter Profile Photo

This is like asking: What % of ball touches in a football game result in a goal? It's a bad measure of success, not every pass is a shot on goal and it can need many passes to create a chance. Research works similarly. Most papers aren't breakthroughs and some are misses.

Jamison Polackwich (@rjpolackwich) 's Twitter Profile Photo

Why does the horn in basketball have to be so loud like that? You can get the message across without being so harsh I feel like

GonzoML (@che_shr_cat) 's Twitter Profile Photo

1/ Transformers don't count like computers. We assume they have hidden "registers" to track variables. We were wrong. New research by Anthropic reverse-engineered Claude 3.5 Haiku and found it works with 6D helical manifolds. It's geometry, not math. 🧵

1/
Transformers don't count like computers. We assume they have hidden "registers" to track variables. We were wrong.

New research by <a href="/AnthropicAI/">Anthropic</a> reverse-engineered Claude 3.5 Haiku and found it works with 6D helical manifolds.

It's geometry, not math. 🧵