paper-with-me

Papers

An SDE for Modeling SAM: Theory and Insights

2023-01-19 · Enea Monzio Compagnoni, Luca Biggio, Antonio Orvieto, Frank Norbert Proske, Hans Kersting, Aurelien Lucchi

We study the SAM (Sharpness-Aware Minimization) optimizer which has recently attracted a lot of interest due to its increased performance over more classical variants of stochastic gradient descent. Our main contribution is the derivation of continuous-time models (in the form of SDEs) for SAM and two of its variants, both for the full-batch and mini-batch settings. We demonstrate that these SDEs are rigorous approximations of the real discrete-time algorithms (in a weak sense, scaling linearly with the learning rate). Using these models, we then offer an explanation of why SAM prefers flat minima over sharp ones~--~by showing that it minimizes an implicitly regularized loss with a Hessian-dependent noise structure. Finally, we prove that SAM is attracted to saddle points under some realistic conditions. Our theoretical results are supported by detailed experiments.

📄 PDF Abstract BibTeX arXiv:2301.08203

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Function Alignment: A New Theory of Mind and Intelligence, Part I: Foundations

2025-03-27 · Gus G. Xia

This paper introduces function alignment, a novel theory of mind and intelligence that is both intuitively compelling and structurally grounded. It explicitly models how meaning, interpretation, and analogy emerge from i…

On the Computation and Applications of Large Dense Partial Correlation Networks

2019-03-17 · Keith Dillon

While sparse inverse covariance matrices are very popular for modeling network connectivity, the value of the dense solution is often overlooked. In fact the L2-regularized solution has deep connections to a number of im…

Dimensionality ReductionModel SelectionUncertainty Quantification

Position: A Dynamical Systems Perspective is Needed to Advance Time Series Modeling

2026-02-18 · Daniel Durstewitz, Christoph Jürgen Hemmer, Florian Hess, Charlotte Ricarda Doll 외 arxiv

Time series (TS) modeling has come a long way from early statistical, mainly linear, approaches to the current trend in TS foundation models. With a lot of hype and industrial demand in this field, it is not always clear…

Conceptive Artificial Intelligence: Insights from design theory

2014-04-02 · Akin Osman Kazakci

The current paper offers a perspective on what we term conceptive intelligence - the capacity of an agent to continuously think of new object definitions (tasks, problems, physical systems, etc.) and to look for methods …

Seeing Through Risk: A Symbolic Approximation of Prospect Theory

2025-04-20 · Ali Arslan Yousaf, Umair Rehman, Muhammad Umair Danish

We propose a novel symbolic modeling framework for decision-making under risk that merges interpretability with the core insights of Prospect Theory. Our approach replaces opaque utility curves and probability weighting …

Decision Making