Google opened the doors to its Machine Learning Crash Course, which had already been taken by more than 18, 000 Googlers, in March 2018. This free course forms the starting point for anyone to learn about and practice ML concepts and comprises 15 hours of material, including instructional videos, interactive visualizations and exercises

While there's many and great Machine Learning courses, the intended audience varies but amongst the beginner friendly ones this one gets the crown. Although tackling it requires knowledge in a few things, namely Numpy and Pandas, there's two very quick onboarding tutorials available on those topics too. In any case only a basic understanding is necessary.

It takes a holisitc approach and aims to equip the students with the answers to the following essential questions:

How does machine learning differ from traditional programming?

What is loss, and how do I measure it?

How does gradient descent work?

How do I determine whether my model is effective?

How do I represent my data so that a program can learn from it?

How do I build a deep neural network?

Specifically it's objectives are three :

Learn to use ML as a tool to reduce the time you spend programming:

For any given problem you can come up after weeks of hard work with a reasonable program, or you can use an off-the-shelf machine learning tool, feed it some examples, and get a more reliable program in a small fraction of the time.

Second, it will allow you to customize your products, making them better for specific groups of people.:

Suppose I produced my English spelling corrector by writing code by hand, and it was so successful that I wanted to have versions in the 100 most popular languages. I would have to start almost from scratch for each language, and it would take years of effort.

But if I built it using machine learning, then moving to another language, to a first approximation, means just collecting data in that language and feeding it into the exact same machine learning model.

And third, machine learning lets you solve problems that you, as a programmer, have no idea how to do by hand:

As a human being, I have the ability to recognize my friends' faces and understand their speech, but I do all of this subconsciously.

So if you asked me to write down a program to do it, I'd be completely baffled.But these are tasks that machine learning algorithms do very well;I don't need to tell the algorithm what to do, I only need to show the algorithm lots of examples, and from that the task can be solved.

As such the complete syllabus that teaches these concepts is as follows:

Introduction to Machine Learning Recognize the practical benefits of mastering machine learning and understand the philosophy behind machine learning

Framing How to frame a task as a machine learning problem. The module covers many of the basic vocabulary terms shared across a wide range of machine learning (ML) methods.

Descending into ML This module explores linear regression before laying the groundwork for a machine learning approach to linear regression.

Refresh your memory on line fitting. Relate weights and biases in machine learning to slope and offset in line fitting and understand "loss" in general and squared loss in particular.

Reducing Loss The meaning of "loss"; to train a model, we need a good way to reduce the model’s loss.

Introduction to TensorFlow Learn enough about NumPy and pandas to understand tf. keras code (the TensorFlow variant of the open-source Keras API).

Generalization Generalization refers to your model's ability to adapt properly to new, previously unseen data, drawn from the same distribution as the one used to create the model.

Training and Test Sets Examine the benefits of dividing a data set into a training set and a test set.

Validation Set: Check Your Intuition Understand the importance of a validation set in a partitioning scheme.

Representation A machine learning model can't directly see, hear, or sense input examples. Instead, you must create a representation of the data to provide the model with a useful vantage point into the data's key qualities. That is, in order to train a model, you must choose the set of features that best represent the data.

Feature Crosses A feature cross is a synthetic feature formed by multiplying (crossing) two or more features. Crossing combinations of features can provide predictive abilities beyond what those features can provide individually.

Regularization for Simplicity Regularization means penalizing the complexity of a model to reduce overfitting.

Logistic Regression Instead of predicting exactly 0 or 1, logistic regression generates a probability—a value between 0 and 1, exclusive.

Classification This module shows how logistic regression can be used for classification tasks, and explores how to evaluate the effectiveness of classification models.

Regularization for Sparsity This module focuses on the special requirements for models learned on feature vectors that have many dimensions.

Neural Networks Neural networks are a more sophisticated version of feature crosses. In essence, neural networks learn the appropriate feature crosses for you.

Embeddings An embedding is a relatively low-dimensional space into which you can translate high-dimensional vectors.

Static vs. Dynamic Training

Static vs. Dynamic Inference

Data Dependencies

Fairness

Finally there's practical examples in which you must use your debugging skills:

ML Systems in the Real World: Cancer Prediction debug a real-world ML problem related to cancer prediction.

ML Systems in the Real World: Literature debug a real-world ML problem related to 18th century literature.

ML Systems in the Real World: Guidelines This lesson summarizes the guidelines learned from these real-world examples.

To sum it up, it's the perfect course for total beginners that certainly deserves a second mention in order to inform those not already aware of it.

Gordon Moore, the co-founder of Intel who pioneered microprocessor technology and whose name is known for Moore's Law, passed away on March 24, 2023. The announcement of his death was made jointly by [ ... ]

In a very interesting webinar streamed live as part of the recent JConference, Antoine Sabot-Durand talked about "hostility" between J2EE/Jakarta and Spring and the differences between them from [ ... ]