Machine Learning
285 views
Overfitting
Quick Definition
When a model learns training data noise and fails on new data
Full Definition
A modeling error where a model learns noise in training data, negatively impacting performance on new data.
Examples
high variance, poor generalization, complex models
Related Terms
regularization
cross-validation
underfitting
More Machine Learning Terms
Bias-Variance Tradeoff
The tension between model simplicity and complexity errors
Data Augmentation
Artificially increasing training data through modified copies
Support Vector Machine
Finding optimal hyperplane to separate classes with max margin
Dropout
Regularization randomly deactivating neurons during training
Contrastive Learning
Self-supervised learning by contrasting data pairs
Hyperparameter Tuning
Finding optimal configuration values before training