paper-with-me

홈 › Papers

Symmetry constrained machine learning

2018-11-16 · Doron L. Bergman

Symmetry, a central concept in understanding the laws of nature, has been used for centuries in physics, mathematics, and chemistry, to help make mathematical models tractable. Yet, despite its power, symmetry has not been used extensively in machine learning, until rather recently. In this article we show a general way to incorporate symmetries into machine learning models. We demonstrate this with a detailed analysis on a rather simple real world machine learning system - a neural network for classifying handwritten digits, lacking bias terms for every neuron. We demonstrate that ignoring symmetries can have dire over-fitting consequences, and that incorporating symmetry into the model reduces over-fitting, while at the same time reducing complexity, ultimately requiring less training data, and taking less time and resources to train.

📄 PDF Abstract BibTeX arXiv:1811.07051

Code (0)

등록된 구현이 없습니다.

Tasks

BIG-bench Machine Learning

Similar Papers 제목 키워드 기반

How unconstrained machine-learning models learn physical symmetries

2026-03-25 · Michelangelo Domina, Joseph William Abbott, Paolo Pegolo, Filippo Bigi 외 arxiv

The requirement of generating predictions that exactly fulfill the fundamental symmetry of the corresponding physical quantities has profoundly shaped the development of machine-learning models for physical simulations. …

Physical SimulationsGraph Neural NetworkData AugmentationPoint Clouds

Symmetry Induces Structure and Constraint of Learning

2023-09-29 · Liu Ziyin

Due to common architecture designs, symmetries exist extensively in contemporary neural networks. In this work, we unveil the importance of the loss function symmetries in affecting, if not deciding, the learning behavio…

Learning Broken Symmetries with Approximate Invariance

2024-12-25 · Seth Nabat, Aishik Ghosh, Edmund Witkowski, Gregor Kasieczka 외

Recognizing symmetries in data allows for significant boosts in neural network training, which is especially important where training data are limited. In many cases, however, the exact underlying symmetry is present onl…

Data Augmentation

Discovery of 2D Materials via Symmetry-Constrained Diffusion Model

2024-12-24 · Shihang Xu, Shibing Chu, Rami Mrad, Zhejun Zhang 외

Generative model for 2D materials has shown significant promise in accelerating the material discovery process. The stability and performance of these materials are strongly influenced by their underlying symmetry. Howev…

Optimizing over trained GNNs via symmetry breaking

2023-09-21 · NeurIPS 2023 11

Optimization over trained machine learning models has applications including: verification, minimizing neural acquisition functions, and integrating a trained surrogate into a larger decision-making problem. This paper f…