Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss

Kaidi Cao, Colin Wei, Adrien Gaidon, +2

arXiv (Cornell University)

Semi-supervised LearningData AugmentationEngineeringMachine LearningData Science +13

Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks

Yuanzhi Li, Colin Wei, Tengyu Ma

arXiv (Cornell University)

EngineeringMachine LearningAutoencodersRegularization EffectTraining Neural Networks +15

GPT-4o System Card

OpenAI, A. M. Hurst, Adam Lerer, +364

arXiv (Cornell University)

Artificial IntelligenceEngineeringMachine LearningSpeech RecognitionNatural Language Processing +16

Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data

Colin Wei, Kendrick Shen, Yining Chen, +1

arXiv (Cornell University)

Artificial IntelligenceTheoretical AnalysisSelf-training AlgorithmsEngineeringMachine Learning +12

On the Margin Theory of Feedforward Neural Networks

Colin Wei, Jason D. Lee, Qiang Liu, +1

arXiv (Cornell University)

Model OptimizationNeural Scaling LawSupport Vector MachineEngineeringMachine Learning +11

Regularization Matters: Generalization and Optimization of Neural Nets v.s. their Induced Kernel

Colin Wei, Jason D. Lee, Qiang Liu, +1

arXiv (Cornell University)

Data-dependent Sample Complexity of Deep Neural Networks via Lipschitz Augmentation

Colin Wei, Tengyu Ma

arXiv (Cornell University)

Lipschitz AugmentationData AugmentationWeight MatricesEngineeringMachine Learning +13

Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks

Yuanzhi Li, Colin Wei, Tengyu Ma

Neural Information Processing Systems

EngineeringMachine LearningAutoencodersRegularization EffectTraining Neural Networks +16

Regularization Matters: Generalization and Optimization of Neural Nets v.s. their Induced Kernel

Colin Wei, Jason D. Lee, Qiang Liu, +1

neural information processing systems

Artificial IntelligenceGradient DescentEngineeringMachine LearningNoisy Gradient Descent +15

Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Loss

Jeff Z. HaoChen, Colin Wei, Adrien Gaidon, +1

arXiv (Cornell University)

Artificial IntelligenceGeometric LearningEngineeringMachine LearningConditional Independence +16