Machine Learning Fundamentals Hands-On Workshop

Duration

16 hours

Objective

Introduce Qualcomm’s new graduate employees to machine learning concepts and provide hands-on experience using popular machine learning libraries like H2O, Scikit-Learn, TensorFlow, and PyTorch. The workshop will focus on practical applications of machine learning on well-selected public datasets.

Detailed Outline

Session 1: Introduction to Machine Learning (2 hours)

Overview of machine learning concepts, supervised and unsupervised learning

Introduction to common machine learning tasks: classification, regression, clustering

Discussion on the importance of data preprocessing, feature engineering, and model evaluation

Session 2: Introduction to H2O (2 hours)

Overview of H2O machine learning platform and its features

Hands-on exercise: Exploring H2O, building and evaluating machine learning models on well-selected public datasets

Session 3: Getting Started with Scikit-Learn (2 hours)

Introduction to Scikit-Learn library and its core functionalities

Hands-on exercise: Loading a dataset, data preprocessing, and performing basic classification and regression tasks using Scikit-Learn

Session 4: Advanced Techniques with Scikit-Learn (2 hours)

Feature selection and dimensionality reduction techniques

Model evaluation and hyperparameter tuning

Hands-on exercise: Applying advanced techniques using Scikit-Learn on well-selected public datasets

Session 5: Introduction to TensorFlow (2 hours)

Overview of TensorFlow library and its role in deep learning

Introduction to neural networks and deep learning concepts

Hands-on exercise: Building and training a basic neural network model using TensorFlow on well-selected public datasets

Session 6: Introduction to PyTorch (2 hours)

Introduction to PyTorch library and its advantages in deep learning

Hands-on exercise: Implementing a neural network using PyTorch and training it on well-selected public datasets

Session 8: Theoretical intro about Generative AI (self-attention, transformers, diffusion model, etc.) (2 hours)

Session 7: Advanced Deep Learning Techniques (2 hours)

Introduction to advanced deep learning techniques: convolutional neural networks (CNNs)

Hands-on exercise: Building and training CNN models using TensorFlow or PyTorch on well-selected public datasets

Session 8: Theoretical intro about Generative AI (self-attention, transformers, diffusion model, etc.) (2 hours)