Core Stability

Core stability, broadly defined as the robustness and reliability of a system or model's performance under various conditions, is a central theme across diverse scientific fields. Current research focuses on improving stability in machine learning models (e.g., through regularization techniques, modified optimization algorithms like AdamG, and analyses of Jacobian alignment), in inverse problems (using methods like Wasserstein gradient flows and data-dependent regularization), and in dynamical systems (by mitigating data-induced instability). Understanding and enhancing core stability is crucial for building reliable and trustworthy systems across applications ranging from medical imaging and AI to robotics and control systems.

Papers

November 14, 2022

Stability and Robustness of Distributed Suboptimal Model Predictive Control
Giuseppe Belgioioso, Dominic Liao-McPherson, Mathias Hudoba de Badyn, Nicolas Pelzmann, John Lygeros, Florian Dörfler
Native Robustness Core Stability Model Predictive Control Large Scale Optimization

November 3, 2022

Fairness in Federated Learning via Core-Stability
Bhaskar Ray Chaudhury, Linyi Li, Mintong Kang, Bo Li, Ruta Mehta
Procedural Fairness Core Stability

October 27, 2022

Too Brittle To Touch: Comparing the Stability of Quantization and Distillation Towards Developing Lightweight Low-Resource MT Models
Harshita Diddee, Sandipan Dandapat, Monojit Choudhury, Tanuja Ganu, Kalika Bali
Knowledge Distillation Training Data Low Resource Language Multilingual Model Core Stability Quantization Operator Mutual Distillation Human Touch

October 19, 2022

October 16, 2022

Stability of Accuracy for the Training of DNNs Via the Uniform Doubling Condition
Yitzchak Shmalo
Deep Neural Network Training Data Core Stability Cross Entropy Loss DNN Framework

October 10, 2022

Second-order regression models exhibit progressive sharpening to the edge of stability
Atish Agarwala, Fabian Pedregosa, Jeffrey Pennington
Gradient Descent High Dimensional Extreme Edge Core Stability Hessian Matrix Second Order

October 4, 2022

Tree Mover's Distance: Bridging Graph Metrics and Stability of Graph Neural Networks
Ching-Yao Chuang, Stefanie Jegelka
Graph Neural Network Strong Generalization Machine Learning Model Graph Drawing Core Stability Support Vector Machine Distance Matter Graph Metric Local Computation

October 3, 2022

On Stability and Generalization of Bilevel Optimization Problem
Meng Ding, Mingxi Lei, Yunwen Lei, Di Wang, Jinhui Xu
Strong Generalization Core Stability Bilevel Optimization Hyper Parameter Optimization Strongly Convex High Probability Generalization

October 2, 2022

Ensembling improves stability and power of feature selection for deep learning models
Prashnna K Gyawali, Xiaoxia Liu, James Zou, Zihuai He
Deep Learning Model Feature Selection Real Power Core Stability Feature Importance Model Ensembling

September 30, 2022

Self-Stabilization: The Implicit Bias of Gradient Descent at the Edge of Stability
Alex Damian, Eshaan Nichani, Jason D. Lee
Gradient Descent Extreme Edge Core Stability Implicit Bias Large Batch Automatic Stabilization

September 26, 2022

Learning GFlowNets from partial episodes for improved convergence and stability
Kanika Madan, Jarrid Rector-Brooks, Maksym Korablyov, Emmanuel Bengio, Moksh Jain, Andrei Nica, Tom Bosc, Yoshua Bengio, Nikolay Malkin
Core Stability Convergence Rate Generative Flow Network Energy Based GFlowNets Sparse Reward Environment Uncertain Episode Length GFlowNet Training

September 22, 2022

A Closer Look at Learned Optimization: Stability, Robustness, and Inductive Biases
James Harrison, Luke Metz, Jascha Sohl-Dickstein
Native Robustness Optimization Purpose Inductive Bias Core Stability Black Box Optimization Glance Annotation Better Optimizers Learned Optimizers

September 19, 2022

Stability and Generalization Analysis of Gradient Methods for Shallow Neural Networks
Yunwen Lei, Rong Jin, Yiming Ying
Strong Generalization Gradient Descent Core Stability Gradient Method Shallow Neural Network

September 16, 2022

Stability and Generalization for Markov Chain Stochastic Gradient Methods
Puyu Wang, Yunwen Lei, Yiming Ying, Ding-Xuan Zhou
Strong Generalization Core Stability Empirical Risk Minimization Generalization Bound Markov Chain Gradient Method Optimal Convergence Rate

September 11, 2022

Stability of Syntactic Dialect Classification Over Space and Time
Jonathan Dunn, Sidney Wong
Time Matter Core Stability Deep Space Dialect Datasets Grammar Induction Dialect Classification Syntactic Representation

September 10, 2022

Examining stability of machine learning methods for predicting dementia at early phases of the disease
Sinan Faouri, Mahmood AlBashayreh, Mohammad Azzeh
Machine Learning Feature Selection Core Stability Support Vector Machine Machine Learning Algorithm Common Disease Dementia Related Linguistic Anomaly Early Stage Dementia Risk

September 1, 2022

Optimising 2D Pose Representation: Improve Accuracy, Stability and Generalisability Within Unsupervised 2D-3D Human Pose Estimation
Peter Hardy, Srinandan Dasmahapatra, Hansung Kim
Core Stability Stronger Generalizability 3D Pose 2 Dimensional Human Pose Estimation 2 Dimensional Pose Enhancing Accuracy Lifted Inference Unsupervised 3D Pose

August 21, 2022

Instability and Local Minima in GAN Training with Kernel Discriminators
Evan Becker, Parthe Pandit, Sundeep Rangan, Alyson K. Fletcher
Generative Adversarial Network Generative Modeling Core Stability GAN Training Mode Collapse Local Minimum Kernel Classification Point Model

July 29, 2022

Adaptive Gradient Methods at the Edge of Stability
Jeremy M. Cohen, Behrooz Ghorbani, Shankar Krishnan, Naman Agarwal, Sourabh Medapati, Michal Badura, Daniel Suo, David Cardoze, Zachary Nado, George E. Dahl, Justin Gilmer
Extreme Edge Core Stability Adaptive Gradient Method Stability Threshold Edge Attribute Non Adaptive

Core Stability

Papers

Stability and Robustness of Distributed Suboptimal Model Predictive Control

Fairness in Federated Learning via Core-Stability

Too Brittle To Touch: Comparing the Stability of Quantization and Distillation Towards Developing Lightweight Low-Resource MT Models

Stability of Entropic Wasserstein Barycenters and application to random geometric graphs

Improving Stability of Fine-Tuning Pretrained Language Models via Component-Wise Gradient Norm Clipping

Stability of Accuracy for the Training of DNNs Via the Uniform Doubling Condition

Second-order regression models exhibit progressive sharpening to the edge of stability

Tree Mover's Distance: Bridging Graph Metrics and Stability of Graph Neural Networks

On Stability and Generalization of Bilevel Optimization Problem

Ensembling improves stability and power of feature selection for deep learning models

Self-Stabilization: The Implicit Bias of Gradient Descent at the Edge of Stability

Learning GFlowNets from partial episodes for improved convergence and stability

A Closer Look at Learned Optimization: Stability, Robustness, and Inductive Biases

Stability and Generalization Analysis of Gradient Methods for Shallow Neural Networks

Stability and Generalization for Markov Chain Stochastic Gradient Methods

Stability of Syntactic Dialect Classification Over Space and Time

Examining stability of machine learning methods for predicting dementia at early phases of the disease

Optimising 2D Pose Representation: Improve Accuracy, Stability and Generalisability Within Unsupervised 2D-3D Human Pose Estimation

Instability and Local Minima in GAN Training with Kernel Discriminators

Adaptive Gradient Methods at the Edge of Stability