Imagen de Portada

Dichotomy of Feature Learning and Unlearning: Fast-Slow Analysis on Neural Networks with Stochastic Gradient Descent

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Imai, Shota, Nishiyama, Sota, Imaizumi, Masaaki
Formato:	Preprint
Publicado:	2026
Materias:	Machine Learning Data Analysis, Statistics and Probability
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Type-II Saddles and Probabilistic Stability of Stochastic Gradient Descent
por: Ziyin, Liu, et al.
Publicado: (2023)

Annealed Stein Variational Gradient Descent for Improved Uncertainty Estimation in Full-Waveform Inversion
por: Corrales, Miguel, et al.
Publicado: (2024)

Active Learning with Fully Bayesian Neural Networks for Discontinuous and Nonstationary Data
por: Ziatdinov, Maxim
Publicado: (2024)

Learnability Window in Gated Recurrent Neural Networks
por: Livi, Lorenzo
Publicado: (2025)

Efficient Training of Deep Neural Operator Networks via Randomized Sampling
por: Karumuri, Sharmila, et al.
Publicado: (2024)

A Pipeline for Data-Driven Learning of Topological Features with Applications to Protein Stability Prediction
por: Mishra, Amish, et al.
Publicado: (2024)

Online Learning Approach for Survival Analysis
por: Fernandez, Camila, et al.
Publicado: (2024)

Stochastic Clock Attention for Aligning Continuous and Ordered Sequences
por: Soh, Hyungjoon, et al.
Publicado: (2025)

Temporal Graph Neural Networks for Early Anomaly Detection and Performance Prediction via PV System Monitoring Data
por: Mukherjee, Srijani, et al.
Publicado: (2025)

Scalable Sparse Regression for Model Discovery: The Fast Lane to Insight
por: Golden, Matthew
Publicado: (2024)

Out-of-Sample Hydrocarbon Production Forecasting: Time Series Machine Learning using Productivity Index-Driven Features and Inductive Conformal Prediction
por: Idris, Mohamed Hassan Abdalla, et al.
Publicado: (2025)

Information-theoretic Quantification of High-order Feature Effects in Classification Problems
por: Lazic, Ivan, et al.
Publicado: (2025)

OASIS: A Deep Learning Framework for Universal Spectroscopic Analysis Driven by Novel Loss Functions
por: Young, Chris, et al.
Publicado: (2025)

Stochastic Inference of Plate Bending from Heterogeneous Data: Physics-informed Gaussian Processes via Kirchhoff-Love Theory
por: Kavrakov, Igor, et al.
Publicado: (2024)

Exploring the Limitations of kNN Noisy Feature Detection and Recovery for Self-Driving Labs
por: Shi, Qiuyu, et al.
Publicado: (2025)

Reservoir Static Property Estimation Using Nearest-Neighbor Neural Network
por: Wang, Yuhe
Publicado: (2024)

Comparative Analysis of Machine Learning-Based Imputation Techniques for Air Quality Datasets with High Missing Data Rates
por: Yan, Sen, et al.
Publicado: (2024)

Position: Solve Layerwise Linear Models First to Understand Neural Dynamical Phenomena (Neural Collapse, Emergence, Lazy/Rich Regime, and Grokking)
por: Nam, Yoonsoo, et al.
Publicado: (2025)

From Data to Laws: Neural Discovery of Conservation Laws Without False Positives
por: Ray, Rahul D
Publicado: (2026)

Detail Across Scales: Multi-Scale Enhancement for Full Spectrum Neural Representations
por: Ni, Yuan, et al.
Publicado: (2025)

Neural posterior estimation for scalable and accurate inverse parameter inference in Li-ion batteries
por: Hassanaly, Malik, et al.
Publicado: (2026)

Dreaming Learning
por: Londei, Alessandro, et al.
Publicado: (2024)

Neural Network Methods for Radiation Detectors and Imaging
por: Lin, S., et al.
Publicado: (2023)

High-Dimensional Limit of Stochastic Gradient Flow via Dynamical Mean-Field Theory
por: Nishiyama, Sota, et al.
Publicado: (2026)

Topological Learning in Multi-Class Data Sets
por: Griffin, Christopher, et al.
Publicado: (2023)

Learning thermodynamically constrained equations of state with uncertainty
por: Sharma, Himanshu, et al.
Publicado: (2023)

GAMMA_FLOW: Guided Analysis of Multi-label spectra by MAtrix Factorization for Lightweight Operational Workflows
por: Rädle, Viola, et al.
Publicado: (2025)

Learning effective good variables from physical data
por: Barletta, Giulio, et al.
Publicado: (2024)

Active Learning of Molecular Data for Task-Specific Objectives
por: Ghosh, Kunal, et al.
Publicado: (2024)

Simultaneous Dimensionality Reduction: A Data Efficient Approach for Multimodal Representations Learning
por: Abdelaleem, Eslam, et al.
Publicado: (2023)

Tunable correlation retention: A statistical method for generating synthetic data
por: Jävergård, Nicklas, et al.
Publicado: (2024)

Learning from the past: predicting critical transitions with machine learning trained on surrogates of historical data
por: Ma, Zhiqin, et al.
Publicado: (2024)

Learning Complex Physical Regimes via Coverage-oriented Uncertainty Quantification: An application to the Critical Heat Flux
por: Cazzola, Michele, et al.
Publicado: (2026)

Learning to solve Bayesian inverse problems: An amortized variational inference approach using Gaussian and Flow guides
por: Karumuri, Sharmila, et al.
Publicado: (2023)

Thermodynamic bounds on energy use in quasi-static Deep Neural Networks
por: Tkachenko, Alexei V.
Publicado: (2025)

Statistical limits of correlation detection in trees
por: Ganassali, Luca, et al.
Publicado: (2022)

DPGIIL: Dirichlet Process-Deep Generative Model-Integrated Incremental Learning for Clustering in Transmissibility-based Online Structural Anomaly Detection
por: Mei, Lin-Feng, et al.
Publicado: (2024)

Short-Term Forecasting of Energy Production and Consumption Using Extreme Learning Machine: A Comprehensive MIMO based ELM Approach
por: Voyant, Cyril, et al.
Publicado: (2025)

Landscape computations for the edge of chaos in nonlinear dynamical systems
por: Nakata, Motoki, et al.
Publicado: (2025)

Toward Dynamic Stability Assessment of Power Grid Topologies using Graph Neural Networks
por: Nauck, Christian, et al.
Publicado: (2022)