🚨Higher-order combinatorial models in TDL are notoriously slow and resource-hungry. Can we do better? Introducing: 🚀 𝐇𝐎𝐏𝐒𝐄: A Scalable Higher-Order Positional and Structural Encoder for Combinatorial Representations 🚀 📝 arXiv: arxiv.org/abs/2505.15405 🧵 (1/6)

Been waiting for this one😼! A beautiful, digestible guide to non-Euclidean ML — gorgeous visuals, clean tables, and clear explanations of topology, algebra, and geometry. Dive in 👇

Today's library addition: A new (2024) history book on how we arrived at the current notation for vectors and general tensors, with an early nod to the quaternion approach and how things could have gone another route.

Psychologists have posited hundreds of cognitive biases over the years. A fascinating paper argues that they all boil down to one of a handful of fundamental beliefs coupled with confirmation bias. [Link below.]

Our latest work on understanding the behavior of bounded agents in more naturalistic tasks with inverse modeling: Revisiting Cost Functions in Sensorimotor Decision-Making with Tobias Niehues & Dominik Straub #CCN2025 more at bsky.app/profile/tobnie…

Why can we decode visual mental contents from primary visual cortex when there’s little evidence for sustained spiking in this part of the brain? Today at #CCN25 poster C104, Noa will present a tentative answer from her work with attractor models. 2025.ccneuro.org/abstract_pdf/K…

Physics of the Riemann Hypothesis arxiv.org/abs/1101.3116

found new "every programmer should read"; last time it was about floating point arithmetic took me few months to read it but it was worth.

Wasserstein Distributionally Robust Optimization: Theory and Applications in Machine Learning arxiv.org/abs/1908.08729

"A geometric modeling of Occam’s razor in deep learning" Open Access PDF: link.springer.com/article/10.100…

Y’all! Ece just published her FIRST first author paper. This one is fun ⬇️

1/2 In 1998, Fields medalist Steve Smale posed 18 math problems for the next century, including: "What are the limits of intelligence, both artificial and human?" Yuan Yao (HKUST) said this question inspired him the most, at the Simons Institute. Video: simons.berkeley.edu/talks/yuan-yao…

Social inequity disrupts reward-based learning nature.com/articles/s4427…

Can ancient wisdom promote moral behavior in LLMs? Published with new data! Contemplative insights improve alignment (d = .96) AND boost cooperation + joint reward on the iterated Prisoner’s dilemma task (d = 7+) using GPT-4o + 4.1 Here’s the basic logic 1/x

Information geometry relies on differential geometry: 3 recommended beginner books to get started: - easy intro to tensors from scratch (Muhlich) - 👍👍 short intro to Riemannian geometry (Godinho & Natario) - intro to symplectic and contact structures (McInerney)

Greater stomach-brain connection is associated with worse mental health. This is opposite to the belief that being ‘more in tune with your body’ improves mental health.

Karl Popper famously argued that psychoanalysis is not falsifiable. (“If a patient resists, it’s a defense; if they agree, it’s a confirmation. Either way, theory wins.”) Because of this, he claimed that psychoanalysis is a metaphysics, instead of a science.

Changing your map’s resolution can change your conclusions. It’s called the Support Effect. And it distorts everything from poverty estimates to climate models. Here’s how it works:

"Taxicab Geometry, an Adventure in Non-Euclidean Geometry" by Eugene F. Krause.

Theoretical neuroscience has room to grow — a Comment article by Ann Kennedy nature.com/articles/s4158…