This course is made of four lectures.

Lecture 1: Machine Learning

In this section, we will dive into the common terminology and algorithms that the ML community is currently using and will prepare the learner with a strong knowledge of the field for the next sections. We will also briefly discuss what the benefits of these technologies are, and what the level of automation to impact the product we can expect from them.

Lecture 2: Data Management

Brief description and best practices in order to access datasets, availability and challenges, sources and tools for data management, the recommended ways to collect data in case this is not available online and we need to curate our own dataset, and how this data can be managed and curated to successfully train ML models.

Lecture 3: Train and Evaluation

In this section, we will review the main framework and hardware required to train and evaluate ML models, as well as recommended techniques to follow when training these complex architectures such as data processing techniques, understanding concepts like over-fitting and under-fitting, when data augmentation is needed, common mistakes and solutions to optimize the performance of these models, as well as ways of evaluating and monitor your model training.

Lecture 4: Productize it

In this section, a review of the previous sections will be given but this time keeping in mind the product development process by providing the learner the tools and concepts in order to be able to measure the business impact, benefits and considerations when launching a product with AI/ML features.

Last updated