Regularization Courses and Certifications

Learn common regularization techniques, implement them in Python, and use them to improve machine learning performance on new data.

• Codecademy
• 2 hours
• Self-Paced
• Free Trial Available

StatQuest with Josh Starmer

Regularization Part 1 - Ridge (L2) Regression

Understand how ridge regression reduces overfitting and variance, selects its penalty with cross-validation, and solves regression problems with limited data.

• YouTube
• 20 minutes
• Self-Paced
• Free Video

Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization

63547 ratings at Coursera

Open the deep learning black box: apply initialization, L2 and dropout regularization, gradient checking, batch normalization, hyperparameter tuning, and optimizers like Momentum, RMSprop and Adam in TensorFlow.

• Coursera
• 3 weeks, 8 hours a week
• Self-Paced
• Paid Course

Fixing Classical Models – Diagnosis & Regularization

Diagnose and improve weak classical machine learning models by fixing evaluation, tuning regularization and tree capacity, and applying early stopping to reduce overfitting.

• CodeSignal
• 1 hour
• Self-Paced
• Free Certificate

Mastering Neural Networks and Model Regularization

Build neural networks from scratch and in PyTorch: perceptrons, back-propagation, L1/L2 and dropout regularization, and CNNs for MNIST images and audio spectrograms.

• Coursera
• 5 weeks, 3 hours a week
• Self-Paced
• Paid Course

Steve Brunton

Robust, Interpretable Statistical Models - Sparse Regression with the LASSO

Learn how LASSO sparse regression builds interpretable models with few variables and reduces overfitting.

• YouTube
• 28 minutes
• Self-Paced
• Free Video

DeepLearning.AI;  Stanford

Supervised Machine Learning: Regression and Classification

32923 ratings at Coursera

Build and train supervised learning models in Python with NumPy and scikit-learn: linear and polynomial regression, feature scaling, logistic regression, and regularization to curb overfitting.

• Coursera
• 3 weeks, 11 hours a week
• Self-Paced
• Paid Course

Introduction to Deep Learning

1 rating at Udacity

Learn when deep learning is the right tool, then build the theory—error functions, gradient descent, backpropagation—into PyTorch code and an MNIST digit classifier.

• Udacity
• 20 hours
• Self-Paced
• Paid Course

Optimizing AI Strategies

Optimize AI models for trading through hyperparameter tuning, regularization, evaluation, and deployment considerations.

• Udacity
• 17 hours
• Self-Paced
• Paid Course

Data Science Courses

Deep Learning: Regularization and Neural Network Normalization Techniques - Lecture 7

Dive into advanced deep learning concepts including layer normalization, FRN, TLU, and Keras implementation while mastering essential regularization techniques and gradient management.

• YouTube
• 52 minutes
• Self-Paced
• Free Video

Data Science Courses

Deep Learning - Regularization Techniques - Lecture 4

Dive into advanced regularization techniques for deep learning, covering weight decay, early stopping, and noise injection methods to enhance model performance and prevent overfitting.

• YouTube
• 1 hour 18 minutes
• Self-Paced
• Free Video

UofU Data Science

Nonlinear Regression and Regularization in Data Mining - Spring 2023

Master nonlinear regression concepts, from polynomial mappings to regularization techniques, while learning practical implementation in Python and understanding cross-validation for model optimization.

• YouTube
• 1 hour 18 minutes
• Self-Paced
• Free Video

Diving Deep into Regression

Practice polynomial and regularized regression—ridge, lasso, and elastic net—through prediction and model-comparison exercises.

• CodeSignal
• 3 hours
• Self-Paced
• Free Certificate

PyTorch Techniques for Model Optimization

Optimize PyTorch training with checkpointing, mini-batches, learning-rate scheduling, regularization, and dropout.

• CodeSignal
• 3 hours
• Self-Paced
• Free Certificate

TensorFlow Techniques for Model Optimization

Optimize TensorFlow models with regularization, dropout, callbacks, and Keras wrappers connecting TensorFlow to scikit-learn.

• CodeSignal
• 3 hours
• Self-Paced
• Free Certificate