arXiv:2502.13024v2 Announce Type: replace
Abstract: Classification models play a central role in data-driven decision-making applications such as medical diagnosis, recommendation systems, and risk assessment. Traditional performance metrics, such as …
arXiv:2505.24275v2 Announce Type: replace
Abstract: We propose GradPower, a lightweight gradient-transformation technique for accelerating language model pre-training. Given a gradient vector $g=(g_i)_i$, GradPower first applies the elementwise sign-p…
arXiv:2601.16514v2 Announce Type: replace
Abstract: Understanding why Transformers perform so well remains challenging due to their non-convex optimization landscape. In this work, we analyze a shallow Transformer with $m$ independent heads trained by…
arXiv:2210.13277v4 Announce Type: replace
Abstract: In distributed optimization, a large number of machines alternate between local computations and communication with a coordinating server. Communication, which can be slow and costly, is the main bot…
arXiv:2506.07134v3 Announce Type: replace
Abstract: We introduce Reliable Policy Iteration (RPI) and Conservative RPI (CRPI), variants of Policy Iteration (PI) and Conservative PI (CPI), that retain tabular guarantees under function approximation. RPI…
arXiv:2604.01406v1 Announce Type: cross
Abstract: We study the problem of identifying an optimal coupling between input-output distributional data generated by a causal dynamical system. The coupling is required to satisfy prescribed marginal distribu…
arXiv:2604.01472v1 Announce Type: cross
Abstract: The Muon optimizer has received considerable attention for its strong performance in training large language models, yet the design principle behind its matrix-gradient orthogonalization remains largel…
arXiv:2604.01525v1 Announce Type: cross
Abstract: We give a short linear–algebraic proof of the inequality \[ \|x\|_1\,\|x\|_\infty \le \frac{1+\sqrt{p}}{2}\,\|x\|_2^2, \] valid for every \(x\in\mathbb{R}^p\). This inequality relates three fundamenta…
arXiv:2604.01606v1 Announce Type: cross
Abstract: Optimization over the space of probability measures endowed with the Wasserstein-2 geometry is central to modern machine learning and mean-field modeling. However, traditional methods relying on full W…
arXiv:2506.15387v2 Announce Type: replace-cross
Abstract: We propose two variants of the Primal Dual Hybrid Gradient (PDHG) algorithm for saddle point problems with block decomposable duals, hereafter called Multi-Timescale PDHG (MT-PDHG) and its acce…