Generalization Performance

Generalization performance in machine learning focuses on a model's ability to accurately predict outcomes on unseen data, a crucial aspect for real-world applications. Current research investigates this through various lenses, including mitigating overfitting in self-supervised and federated learning, improving the robustness of models to out-of-distribution data (e.g., using dropout or orthogonal regularization), and enhancing the efficiency of fine-tuning large pre-trained models (e.g., via low-rank adaptation). Understanding and improving generalization is vital for building reliable and adaptable AI systems across diverse domains, impacting fields from image recognition and natural language processing to control of biological neural networks.

Papers

November 23, 2023

Subnetwork Ensembles
Tim Whitaker
Deep Neural Network Generalization Performance Neural Network Ensemble Low Cost Ensemble

November 21, 2023

November 15, 2023

Predicting generalization performance with correctness discriminators
Yuekun Yao, Alexander Koller
Generalization Performance Sequence to Sequence Multidimensional Local Precision Rate Unseen Data Correctness Check

November 13, 2023

A PAC-Bayesian Perspective on the Interpolating Information Criterion
Liam Hodgkinson, Chris van der Heide, Robert Salomone, Fred Roosta, Michael W. Mahoney
Generalization Performance Implicit Regularization PAC Bayesian Bias Variance Information Criterion Empirical Neural Tangent Kernel

November 10, 2023

Aggregation Weighting of Federated Learning via Generalization Bound Estimation
Mingwei Xu, Xiaofeng Cao, Ivor W. Tsang, James T. Kwok
Generalization Performance Generalization Bound Generalization Error Bound Distributional Robustness Weight Aggregation

November 8, 2023

Data Factors for Better Compositional Generalization
Xiang Zhou, Yichen Jiang, Mohit Bansal
Strong Generalization Generalization Performance Compositional Generalization Generalization Problem Characterization Data Data Complexity

November 3, 2023

Obtaining Explainable Classification Models using Distributionally Robust Optimization
Sanjeeb Dash, Soumyadip Ghosh, Joao Goncalves, Mark S. Squillante
High Explainability Generalization Performance Model Explanation Distributionally Robust Optimization Explainable Classification

October 25, 2023

Grokking in Linear Estimators -- A Solvable Model that Groks without Understanding
Noam Levi, Alon Beck, Yohai Bar-Sinai
Generalization Performance Deeper Network Linear Estimator Grokking Phenomenon Linear Network Nonlinear Activation

October 24, 2023

What Algorithms can Transformers Learn? A Study in Length Generalization
Hattie Zhou, Arwen Bradley, Etai Littwin, Noam Razin, Omid Saremi, Josh Susskind, Samy Bengio, Preetum Nakkiran
Transformer Megatron Decepticons Practical Algorithm Study Feature Generalization Performance Compositional Generalization Length Generalization Robotic Assembly Sequence Planning

October 19, 2023

October 15, 2023

Auto-LfD: Towards Closing the Loop for Learning from Demonstrations
Shaokang Wu, Yijin Wang, Yanlong Huang
LeArning Abstract Generalization Performance Noisy Demonstration Dynamic Movement Primitive Learning From Demonstration Peg in Hole Task Control Input Inference

October 14, 2023

Does CLIP's Generalization Performance Mainly Stem from High Train-Test Similarity?
Prasanna Mayilvahanan, Thaddäus Wiedemer, Evgenia Rusak, Matthias Bethge, Wieland Brendel
Supervised ImageNet Generalization Performance Single CLIP Distribution Generalization Generalizable Representation STEM Education Train Test

October 12, 2023

Effects of Human Adversarial and Affable Samples on BERT Generalization
Aparna Elangovan, Jiayuan He, Yuan Li, Karin Verspoor
Training Data Adversarial Example Mixed Effect BERT Model Generalization Performance Stronger Generalizability Adversarial Text Adversarial Information Novel Sample

October 11, 2023

Quantifying Agent Interaction in Multi-agent Reinforcement Learning for Cost-efficient Generalization
Yuxin Chen, Chen Tang, Ran Tian, Chenran Li, Jinning Li, Masayoshi Tomizuka, Wei Zhan
Strong Generalization Multi Agent Reinforcement Learning Generalization Performance Agent Interaction

October 8, 2023

ZSC-Eval: An Evaluation Toolkit and Benchmark for Multi-agent Zero-shot Coordination
Xihuai Wang, Shao Zhang, Wenhao Zhang, Wentao Dong, Jingxiao Chen, Ying Wen, Weinan Zhang
New Benchmark Generalization Performance Cooperative Multi Agent Reinforcement Learning G Eval Zero Shot Coordination

October 1, 2023

Mind the Gap: Federated Learning Broadens Domain Generalization in Diagnostic AI Models
Soroosh Tayebi Arasteh, Christiane Kuhl, Marwin-Jonathan Saehn, Peter Isfort, Daniel Truhn, Sven Nebelung
Federated Learning Generalization Performance Human Mind Data Diversity Domain Performance Diagnostic Accuracy Dataset Diversity

September 30, 2023

On Memorization and Privacy Risks of Sharpness Aware Minimization
Young In Kim, Pratiksha Agrawal, Johannes O. Royset, Rajiv Khanna
Generalization Performance Sharpness Aware Minimization Privacy Risk Limited Memorization Privacy Utility Vanilla Gradient Descent

September 28, 2023

Optimal Nonlinearities Improve Generalization Performance of Random Features
Samet Demir, Zafer Doğan
Generalization Performance Activation Function Random Feature Nonlinear Function Non Monotonic Activation Function