Advanced Machine Learning

For organizations looking for a deeper dive, our five-day in-depth course offers hands-on training and real-world sector specific case studies, presented online or on-premise upon request. The course covers the following topics

Introducing A.I: A brief history and defining core concepts, overview of machine learning pipeline, and examples of real-world applications.

Data exploration: Data collection and visualization, as well as explorative analytics.

Data preprocessing: Addressing data quality problems and performing data cleaning, dimensionality reduction, and feature engineering.

ML Taxonomy: Overview of types of machine learning such as supervised, unsupervised, semi-supervised, and reinforcement learning. Differences between deep learning and traditional machine learning.

Supervised learning: Discussion of popular traditional supervised machine learning models including logistic regression, support vector machines, decision trees, etc.

Model training: Discussing overfitting vs. underfitting, hyperparameter tuning, dataset splitting and leakage, optimization strategies, and evaluating model performance.

Unsupervised learning: Discussion of popular unsupervised learning algorithms including KNN, K-means, hierarchical clustering, GMM, etc.

Sciket-Learn: Introducing scikit-learn and the scikit-learn API. Discussing scikit-learn pitfalls and best practices as well as examples of scikit-learn implementations.

Deep learning: Introducing the deep learning revolution, discussion of perceptrons and neural networks, and an overview of RNN, CNN, and autoencoder architectures.

Deep frameworks: Understanding PyTorch and PyTorch Lightning, as well as a basic overview of JAX, TensorFlow, and Keras.

End-to-end design: Dataset design, model and framework selection, as well as pretraining and transfer learning.

Practical problem: Working through design and implementation for a guided problem.