We continue the investigation into the power of smaller Transformer-base...
Language models (LMs) are powerful tools for natural language processing...
Artificial intelligence (AI) researchers have been developing and refini...
We analyze the hit-and-run algorithm for sampling uniformly from an isot...
We consider a variant of the classical notion of noise on the Boolean
hy...
We study the probability of Boolean functions with small max influence t...
We propose a synthetic task, LEGO (Learning Equality and Group Operation...
We give an alternative, simple method to prove isoperimetric inequalitie...
Two recent and seemingly-unrelated techniques for proving mixing bounds ...
We study the extent to which wide neural networks may be approximated by...
We make the first steps towards generalizing the theory of stochastic bl...
We reduce the problem of proving a "Boolean Unique Games Conjecture" (wi...
In 1988, Eric B. Baum showed that two-layers neural networks with thresh...
We develop a new technique for proving concentration inequalities which
...
We strengthen several classical inequalities concerning the influences o...
Answering a question of Abbasi-Zadeh, Bansal, Guruganesh, Nikolov, Schwa...
Existing depth separation results for constant-depth networks essentiall...
We prove stability estimates for the Shannon-Stam inequality (also known...
We introduce a new method for obtaining quantitative convergence rates f...
We consider the adversarial convex bandit problem and we build the first...
We show that there is a simple (approximately radial) function on ^d,
ex...