Image de couverture de livre

Understanding Mode Connectivity via Parameter Space Symmetry

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhao, Bo, Dehmamy, Nima, Walters, Robin, Yu, Rose
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Latent Space Symmetry Discovery
par: Yang, Jianke, et autres
Publié: (2023)

Symmetry-Informed Governing Equation Discovery
par: Yang, Jianke, et autres
Publié: (2024)

Symmetry in Neural Network Parameter Spaces
par: Zhao, Bo, et autres
Publié: (2025)

AtlasD: Automatic Local Symmetry Discovery
par: Bhat, Manu, et autres
Publié: (2025)

Discovering Symbolic Differential Equations with Symmetry Invariants
par: Yang, Jianke, et autres
Publié: (2025)

Improving Convergence and Generalization Using Parameter Symmetries
par: Zhao, Bo, et autres
Publié: (2023)

Improving Learning to Optimize Using Parameter Symmetries
par: Zamir, Guy, et autres
Publié: (2025)

The Empirical Impact of Neural Parameter Symmetries, or Lack Thereof
par: Lim, Derek, et autres
Publié: (2024)

NRGPT: An Energy-based Alternative for GPT
par: Dehmamy, Nima, et autres
Publié: (2025)

To Augment or Not to Augment? Diagnosing Distributional Symmetry Breaking
par: Lawrence, Hannah, et autres
Publié: (2025)

Discovering Symmetry Breaking in Physical Systems with Relaxed Group Convolution
par: Wang, Rui, et autres
Publié: (2023)

Understanding the Difficulty of Solving Cauchy Problems with PINNs
par: Wang, Tao, et autres
Publié: (2024)

A Survey of Weight Space Learning: Understanding, Representation, and Generation
par: Han, Xiaolong, et autres
Publié: (2026)

Connecting Independently Trained Modes via Layer-Wise Connectivity
par: Tian, Yongding, et autres
Publié: (2025)

Circumventing Backdoor Space via Weight Symmetry
par: Peng, Jie, et autres
Publié: (2025)

Small Models, Smarter Learning: The Power of Joint Task Training
par: Both, Csaba, et autres
Publié: (2025)

Can LLMs Understand Time Series Anomalies?
par: Zhou, Zihao, et autres
Publié: (2024)

Landscaping Linear Mode Connectivity
par: Singh, Sidak Pal, et autres
Publié: (2024)

Proving Linear Mode Connectivity of Neural Networks via Optimal Transport
par: Ferbach, Damien, et autres
Publié: (2023)

Input Space Mode Connectivity in Deep Neural Networks
par: Vrabel, Jakub, et autres
Publié: (2024)

Leveraging Parameter Space Symmetries for Reasoning Skill Transfer in LLMs
par: Horoi, Stefan, et autres
Publié: (2025)

Relaxed Equivariant Graph Neural Networks
par: Hofgard, Elyssa, et autres
Publié: (2024)

Generalized Linear Mode Connectivity for Transformers
par: Theus, Alexander, et autres
Publié: (2025)

Layer-wise Linear Mode Connectivity
par: Adilova, Linara, et autres
Publié: (2023)

Understanding Machine Unlearning Through the Lens of Mode Connectivity
par: Cheng, Jiali, et autres
Publié: (2025)

On Linear Mode Connectivity of Mixture-of-Experts Architectures
par: Tran, Viet-Hoang, et autres
Publié: (2025)

Linear Mode Connectivity in Differentiable Tree Ensembles
par: Kanoh, Ryuichi, et autres
Publié: (2024)

Analyzing the Role of Permutation Invariance in Linear Mode Connectivity
par: Zhan, Keyao, et autres
Publié: (2025)

On Uncertainty Calibration for Equivariant Functions
par: Berman, Edward, et autres
Publié: (2025)

On Universality of Deep Equivariant Networks
par: Pacini, Marco, et autres
Publié: (2025)

Reducing the Sensitivity of Neural Physics Simulators to Mesh Topology via Pretraining
par: Vaska, Nathan, et autres
Publié: (2025)

ParamReL: Learning Parameter Space Representation via Progressively Encoding Bayesian Flow Networks
par: Wu, Zhangkai, et autres
Publié: (2024)

Unveiling Mode Connectivity in Graph Neural Networks
par: Li, Bingheng, et autres
Publié: (2025)

Linear Mode Connectivity in Sparse Neural Networks
par: McDermott, Luke, et autres
Publié: (2023)

Smoothness Errors in Dynamics Models and How to Avoid Them
par: Berman, Edward, et autres
Publié: (2026)

Analysis of Linear Mode Connectivity via Permutation-Based Weight Matching: With Insights into Other Permutation Search Methods
par: Ito, Akira, et autres
Publié: (2024)

MatrixNet: Learning over symmetry groups using learned group representations
par: Laird, Lucas, et autres
Publié: (2025)

Federated Learning over Connected Modes
par: Grinwald, Dennis, et autres
Publié: (2024)

DeepWeightFlow: Re-Basined Flow Matching for Generating Neural Network Weights
par: Gupta, Saumya, et autres
Publié: (2026)

Understanding Hallucinations in Diffusion Models through Mode Interpolation
par: Aithal, Sumukh K, et autres
Publié: (2024)