Machine learning study needs a progression from framing a problem to understanding models and evaluating their behaviour. Google's courses in this selection include a broad crash course, focused concepts and a language-model development sequence. Choose a foundation before moving into architecture, fine-tuning or efficiency simply because those topics sound more advanced.
Table of Contents
Courses to consider first
| Learning goal | Recommended course |
|---|---|
| For deciding whether ML fits a problem | Google Machine Learning Problem Framing: Free Beginner Course |
| For a broad machine-learning foundation | Machine Learning Crash Course |
| For a focused unsupervised-learning topic | Google Introduction to Clustering: Free Machine Learning Course |
| For a language-model development route | Google DeepMind: Build Your Own Small Language Model — Free Course |
For deciding whether ML fits a problem: Problem Framing helps connect a proposed machine-learning task with the outcome it needs to support.
For a broad machine-learning foundation: The Machine Learning Crash Course is the main broad learning route in this collection.
For a focused unsupervised-learning topic: Choose clustering when grouping observations is the concept you need to understand more clearly.
For a language-model development route: The small language-model course connects early model ideas with a training workflow. The related representation, architecture and fine-tuning courses develop distinct parts of that route.
17 courses in this Google selection
-
1. Machine Learning Crash Course
A practical introduction to machine learning using videos, interactive explanations, and exercises.
-
2. Google Machine Learning Problem Framing: Free Beginner Course
Turn an AI idea into a clear goal, model output and success criteria with Google’s free machine learning problem-framing course and written exercises.
-
3. Google Introduction to Clustering: Free Machine Learning Course
Understand k-means, feature scaling, similarity measures and cluster evaluation through Google’s free self-paced clustering course.
-
4. Google Attention Mechanism: Free Deep Learning Course
Understand how attention helps neural networks focus on relevant information in a short, free Google Skills course with a knowledge check.
-
5. Google Transformer Models and BERT: Free NLP Course
Understand Transformer architecture, self-attention and BERT through a free Google course with an overview, recorded walkthrough and quiz.
-
6. Google Create Image Captioning Models: Free Deep Learning Course
Explore how image captioning combines an encoder and decoder, then follow Google’s free overview, model walkthrough and knowledge quiz.
-
7. Google Encoder-Decoder Architecture: Free Deep Learning Course
Learn how encoder-decoder models handle sequence tasks and follow a TensorFlow poetry-generation walkthrough in Google’s free short course.
-
8. Google Introduction to Reliable Deep Learning
Study prediction uncertainty, overconfidence and reliability methods in Google’s free reading-based deep learning course.
-
9. Google MLOps Model Evaluation: Free Agent Platform Course
Study predictive and generative model evaluation, task-specific metrics and evaluation workflows in Google’s free Intermediate MLOps course.
-
10. Google DeepMind: Build Your Own Small Language Model — Free Course
Explore n-grams, Transformer models and the Keras training workflow while connecting your first small language model with a meaningful problem.
-
11. Google DeepMind: Represent Your Language Data — Free Tokenization Course
Explore tokenization, byte pair encoding and embeddings, then connect language-data representation with privacy, documentation and community context.
-
12. Google DeepMind: Design and Train Neural Networks — Free Course
Learn to interpret loss curves, design multilayer perceptrons and investigate generalization through Keras exercises and responsible model reflection.
-
13. Google DeepMind: Discover Transformer Architecture — Free Course
Explore attention, positional embeddings and Transformer components through implementation exercises, architectural limits and community-focused reflection.
-
14. Google DeepMind: Fine-Tune Your Model — Free Course
Compare pre-training and fine-tuning, explore full-parameter adaptation and LoRA, and examine the data and governance choices behind instruction-following models.
-
15. Google DeepMind: Accelerate Your Model — Free Efficiency Course
Explore compute estimates, GPU memory, mixed precision and gradient accumulation while considering the resource and environmental trade-offs of model training.
-
16. Google DeepMind Capstone: Develop Your Model for Real-World Impact — Free Course
Bring dataset preparation, LoRA fine-tuning and evaluation together in a project-focused capstone that connects model development with a defined community need.
-
17. Introduction to AI and Machine Learning on Google Cloud
Explore predictive and generative AI, development options, and the data-to-AI workflow in Google’s introductory course.
Choose your learning route
- Frame a problem and establish the foundations through the crash course.
- Use focused courses to deepen the concepts your project needs, such as clustering, attention or reliable deep learning.
- Follow the language-model sequence from data representation and neural networks toward Transformers, fine-tuning, efficiency and the capstone.
Frequently asked questions
Which course should I take before a language-model project?
The crash course provides broader ML foundations. Within the language-model route, data representation and neural-network concepts are useful before architecture and fine-tuning.
Is the clustering course the same as a general ML introduction?
No. It is a focused topic. Choose the crash course for a broader foundation and clustering when you want to study that particular method.
Questions & discussion
Share a useful question or correction. Comments appear after moderation. Please avoid personal or sensitive information.