My GT dynamical ML group is recruiting PhD students for Fall'25. Interested in ML theory or AI4Science with Math/Phys/CS background?
Plz apply at https://t.co/HuKI7OG087 to either Math, ML, CSE, or ACO PhD program, mention my name & choose Math as home unit.
RT is appreciated!
Experiments and a mathematical model show that bacterial colonies form flows at their edges. Without any preferred chirality, the colonies will collectively break chiral symmetry; this phenomenon is due to a chiral bias in individual #bacteria.
🔗 https://t.co/NtDwGAiKGa
We can compare how frequently astrologers got different numbers of questions correct to how often we'd expect them to get different numbers correct if they were all guessing totally at random with no skill.
The two distributions match very closely.
You can get this result (and many related results) with the Cauchy residue theorem! Integrate around the complex unit circle with 2cos(theta) = z + 1/z.
Remarkable new paper by my postdoc Matt Golden, with a novel algorithm for model discovery in physics. Training that would have taken the age of the universe to complete with prior algorithms can now be achieved in a day. #machinelearning#physics
https://t.co/8d7wNct5xj
This is a nice example of Birdtracks! My favorite two line proof directly follows from one-forms *at any point* being the gradients of trivial scalar fields.
Here's an application of birdtrack notation to prove a curvature tensor identity. It is identical to the [abstract] index notation, but I find the diagrammatic proof easier to follow once I picked up the rules.
This is a solution to exercise [14.10] in Penrose's Road to Reality.
Nature Comms paper: Subtle adversarial image manipulations influence both human and machine perception! We show that adversarial attacks against computer vision models also transfer (weakly) to humans, even when the attack magnitude is small. https://t.co/O7skDZe6zU
This is how democracy fails. A candidate is clearly ineligible for office according to the Constitution, but this fact will be widely ignored by all institutions that matter. Laws are just suggestions; they are only as strong as the decisions of the people who implement them.