Charles H. Martin

18 papers A* 4A 1Journal 13

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	Late-Stage Generalization Collapse in Grokking: Detecting anti-grokking with Weightwatcher. CoRR Hari Prakash, Charles H. Martin
2025	J	`jnl`	Grokking and Generalization Collapse: Insights from HTSR theory. CoRR Hari Prakash, Charles H. Martin
2025	J	`jnl`	SETOL: A Semi-Empirical Theory of (Deep) Learning. CoRR Charles H. Martin, Christopher Hinrichs
2024	J	`jnl`	The recent Physics and Chemistry Nobel Prizes, AI, and the convergence of knowledge fields. Patterns Charles H. Martin, Ganesh Mani
2023	A*	`conf`	Temperature Balancing, Layer-wise Weight Analysis, and Neural Network Training. NeurIPS Yefan Zhou, Tianyu Pang, Keqin Liu, Charles H. Martin, Michael W. Mahoney, Yaoqing Yang
2023	J	`jnl`	Temperature Balancing, Layer-wise Weight Analysis, and Neural Network Training. CoRR Yefan Zhou, Tianyu Pang, Keqin Liu, Charles H. Martin, Michael W. Mahoney, Yaoqing Yang
2023	A*	`conf`	Test Accuracy vs. Generalization Gap: Model Selection in NLP without Accessing Training or Testing Data. KDD Yaoqing Yang, Ryan Theisen, Liam Hodgkinson, Joseph E. Gonzalez, Kannan Ramchandran, Charles H. Martin, Michael W. Mahoney
2022	J	`jnl`	Evaluating natural language processing models with generalization metrics that do not need access to any training or testing data. CoRR Yaoqing Yang, Ryan Theisen, Liam Hodgkinson, Joseph E. Gonzalez, Kannan Ramchandran, Charles H. Martin, Michael W. Mahoney
2021	J	`jnl`	Implicit Self-Regularization in Deep Neural Networks: Evidence from Random Matrix Theory and Implications for Learning. J. Mach. Learn. Res. Charles H. Martin, Michael W. Mahoney
2021	J	`jnl`	Post-mortem on a deep learning contest: a Simpson's paradox and the complementary roles of scale metrics versus shape metrics. CoRR Charles H. Martin, Michael W. Mahoney
2020	A	`conf`	Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks. SDM Charles H. Martin, Michael W. Mahoney
2020	J	`jnl`	Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data. CoRR Charles H. Martin, Tongsu Peng, Michael W. Mahoney
2019	J	`jnl`	Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks. CoRR Charles H. Martin, Michael W. Mahoney
2019	A*	`conf`	Statistical Mechanics Methods for Discovering Knowledge from Modern Production Quality Neural Networks. KDD Charles H. Martin, Michael W. Mahoney
2019	A*	`conf`	Traditional and Heavy Tailed Self Regularization in Neural Network Models. ICML Michael W. Mahoney, Charles H. Martin
2019	J	`jnl`	Traditional and Heavy-Tailed Self Regularization in Neural Network Models. CoRR Charles H. Martin, Michael W. Mahoney
2018	J	`jnl`	Implicit Self-Regularization in Deep Neural Networks: Evidence from Random Matrix Theory and Implications for Learning. CoRR Charles H. Martin, Michael W. Mahoney
2017	J	`jnl`	Rethinking generalization requires revisiting old ideas: statistical mechanics approaches and complex learning behavior. CoRR Charles H. Martin, Michael W. Mahoney