Applied Machine Learning
Design, build, and debug machine learning models for classification and regression tasks using a variety of datasets with Python (Numpy, Scikit, Pyplot). Learn best practices to plan and execute ML development projects whether large or small.
Senior Manager at Apple and Instructor at Stanford University
Andrew Maas is currently at Apple working on data-centric deep learning. He completed a PhD in Computer Science at Stanford in 2015 advised by Andrew Ng and Dan Jurafsky. His dissertation focused on large scale deep learning methods for spoken and written language. Andrew has worked as an engineer and scientific advisor to several startups including Wit.ai, Coursera, and Semantic Machines. Prior to Apple, he built an NLP platform for precise healthcare language as cofounder of Roam Analytics. Additionally he also teaches CS224S: Spoken Language Processing as a visiting lecturer at Stanford University.
Supervised machine learning has emerged as a fundamental tool when building compelling products for a wide range of industries and applications. Machine learning systems performing classification and regression tasks likely currently help daily with tasks including sorting emails, editing photos, interacting with digital assistants, and protecting you from fraud. Developing and delivering production-ready ML systems requires technical understanding of ML models, and effective ML project planning. There are infinite possibilities for how ML systems can integrate into larger products/systems, and a working understanding of ML is critical to design effective, ethical, and maintainable ML-powered products.
In this course, you will develop technical skills by building and debugging widely used supervised ML modeling approaches ranging from robust baselines to state-of-the-art. A range of datasets, problem formulations, and diagnostic techniques will provide hands-on ML engineering experience for overcoming challenges when building ML models. Learn and practice principles for successful ML project planning and execution based on lean startup thinking. These principles and experiences will help to design and execute ML projects large or small.
The course focuses on best practices for efficiently applying supervised machine learning to real world projects. You’ll draw on the research, startup, and big tech experience of instructors and classmates to improve your work with ML – whether you’re already involved in ML projects or just starting the path towards ML engineering.
- ML models and evaluation metrics for regression
- Basic formulation of supervised ML tasks and modeling assumptions
- Formulate and build an ML system for a business challenge
- Models to generate binary, multi-class, and other classification outputs
- Evaluation metrics and diagnostics for classification modeling tasks
- ML problem formulation and project planning steps
- State of the art modeling techniques and best practices to achieve good results on new datasets
- Methods to preprocess and featurize common data types (text, images, audio, tabular)
- Introductory data-centric ML techniques to improve performance of a final model
- Foundation model features and adversarial inputs
- Ethics and bias considerations in ML projects
- Where to go next? How to continue building upon this class in industry or further studies
Co:rise is young but already in a league of its own when compared to other online upskill/career changing courses. Courses are intimate and students are driven. It felt more like an accelerated university level course than on online certificate program in that I learned more in 4 weeks that I have with any other online course. The most unique part is that you get facetime with instructors and mentors who have proven track records in the domain they are teaching.
Taking the Applied Machine Learning course has been an incredible experience. We not only learned tactical skills to approach building state-of-the-art ML models, but also learned important ideas on how to properly setup ML teams, formulate problems, and think about ethics. All of this was supplemented with fireside chats with industry ML practitioners and leaders who talked about their experiences building teams and integrating ML into their products. It's been an amazing 4 weeks, and I'm so happy to have gotten to know the course staff and many of the other students in our cohort. Looking forward to my next co:rise class :)
I was so excited about this class, that I dropped my grad class that I was taking at the same time. One of the things that got me really excited is Andrew’s years in the field allowed him to take complicated concepts and simplify them. When Andrew talked about problem formulation, and running a smaller experiment, it was pivotal in giving me the confidence at work to propose a smaller solution, publish early, talk about our methods. Mentorship with the course team was AMAZING.
You can learn a lot in this course even if you don’t have much prior Python or ML experience as long as you are willing to put in some time on the projects
A crash course in learning the basics to get you started on building your first ML models while using state of the art techniques to further improve performance.
I would definitely recommend the Corise community as it provides a major incentive and community compared to most MOOCs - It's what you make of it, but if you're invested and put a lot in, you will get even more out
Started as a novice to machine learning I get to learn so much in just 4 weeks of course. Starting from basics to training state of the art ML models. The way projects are designed is exceptional and is very helpful in learning quick. Team is helpful and very prompt in responding to any queries.
Anyone involved in machine learning projects seeking to design, plan, and execute projects more effectively
People looking for a technical introduction to applying ML techniques in an industry setting
Software engineers transitioning to machine learning engineering projects
Basic data science with Python (Numpy, Pandas, Pyplot, or similar). Co:rise Python for Machine Learning course or equivalent.
Enough statistics and linear algebra to keep pace with guided scikit-learn ML modeling. At minimum, some experience in statistics with random variables and linear algebra