Review: Machine Learning & AI Specialization with Andrew Ng
As robots continue to take over the world, more and more companies are in search of tech workers with artificial intelligence and machine learning experience. Just kidding about the robots, but AI/ML continues to dominate the job market in technical fields. Being versed in these concepts, even at a high level, gives you a leg up over other people attempting to break into the same roles. If a career in tech sounds interesting to you, we recommend reading on to see if Stanford's Machine Learning Specialization is right for you.
This program will teach you the fundamentals of machine learning and real-world AI applications. It's taught by AI expert Andrew Ng, a renowned researcher at Stanford and groundbreaking work at leading tech companies.
This 3-course program, which should take you around 10 weeks / 2 months to complete, has been rated 4.9 out of 5 by over 4.8 million learners since its launch in 2012. The courses that make up the specialization include:
- Supervised Machine Learning: Regression & Classification
- Advanced Learning Algorithms
- Unsupervised Learning: Recommenders & Reinforcement Learning
Ng covers a wide range of practical machine learning topics, including supervised learning, unsupervised learning, and best practices for AI and machine learning innovation in technical careers.
By the end of the program, you'll have mastered key concepts and gained the skills you need to apply machine learning to real-world problems with confidence.
If you're familiar with product roles, you can use the information in these courses to break in to several careers, including UX researcher, product manager or almost any type of analyst. Putting a machine learning course from Stanford on your resume is definitely a way to step up from the pack.
We’ve recapped the learning objectives from each week to set your expectations for course material. The great part about this program is that you can jump to any course, and any section if it’s interesting to you. For example, if you’re already a product manager and just looking to rock a recommendations algorithm for your product, jump to course 3, week 2. However, you only can get the certificate if you complete all 17 weeks of content.
To audit an individual week-- find the exact course (we've linked them individually here) and click "audit" to save it to your profile. Then open the desired week on the side panel that aligns with our recaps.
Learning Objectives from Week 1: Introduction to Machine Learning
- Understand the fundamental concepts and techniques of machine learning, including supervised and unsupervised learning.
- Begin to write and run Python code in Jupyter notebooks to implement machine learning algorithms and models.
- Know how to define and optimize a regression model using gradient descent, including the implementation and visualization of a cost function.
Learning Objectives from Week 2: Regression with Multiple Input Variables
- Learn various mathematical techniques to make machine learning calculations faster and more efficient.
- Use feature scaling, feature engineering, and polynomial regression to improve the performance of a machine learning model.
- Write code to create a machine learning model that can predict numerical values using linear regression.
Learning Objectives from Week 3: Classification
- Use a specific type of machine learning model called logistic regression to predict one of two outcomes, known as binary classification.
- Write binary classification code to predict outcomes.
- Learn techniques to prevent a machine learning model from being too complex and overfitting to the training data.
Learning Objectives from Week 1: Neural Networks
- Understand the fundamental concepts and components of a neural network, including layers and activations.
- Learn how to use a neural network for image classification, including building a neural network in TensorFlow or regular Python code.
- Learn about advanced techniques for improving the performance of a neural network, such as parallel processing.
Learning Objectives from Week 2: Neural Network Training
- Use TensorFlow to train a neural network on data and understand the importance of different activation functions.
- Understand and implement multiclass classification, including the use of the softmax activation and categorical cross entropy loss function.
- Learn about the difference between multi-label and multiclass classification.
Learning Objectives from Week 3: Advice for Applying Machine Learning
- Understand how to evaluate and improve the performance of a learning algorithm, including techniques such as regularization and error analysis.
- Learn about the iterative process of developing and updating a machine learning model, including techniques like data augmentation and transfer learning.
- Understand the concept of bias and variance and how they apply to neural networks.
- Learn about fairness and ethics in machine learning and how to measure precision and recall when working with imbalanced datasets.
Learning Objectives from Week 4: Decision Trees
- Understand the structure and use of decision trees for making predictions.
- Learn about the impurity metric "entropy" and how it is used in the construction of decision trees.
- Learn about advanced techniques for using decision trees, such as tree ensembles and boosting, and understand when to use decision trees or neural networks.
Learning Objectives from Week 1: Unsupervised Learning
- Implement the k-means clustering algorithm, including the optimization objective, initialization, and centroid update function.
- Understand how to choose the number of clusters for the k-means algorithm and when to use supervised learning versus anomaly detection.
- Implement an anomaly detection system and the function that finds the closest centroids to each point in k-means.
Learning Objectives from Week 2: Recommender Systems
- Learn how to work with collaborative filtering recommender systems in TensorFlow.
- Utilize deep learning content-based filtering using a neural network in TensorFlow.
- Understand ethical considerations in building recommender systems.
Learning Objectives from Week 3: Continued Linear Regression
- Study outliers, inference and testing with in linear regression and variability partitioning.
Learning Objectives from Week 4: Reinforcement Learning
- Understand key concepts in reinforcement learning, including return, state, action, and policy.
- Understand the Bellman equations and the state-action value function.
- Understand continuous state spaces and how to build a deep Q-learning network.
While this is a high-level and beginner friendly program, there is some coding involved with Python. Python is probably one of the most popular coding languages for data science and machine learning, which makes the course extra valuable.
HOWEVER, if you're not looking to code, this may not be the program for you.
Cost and Auditing
The program is only $49/month, and comes with a Linkedin Certificate on behalf of Duke University (remember– this is a prestigious place!!). If you complete the curriculum on the proposed timeline, it should take about 2-3 months, though you could blitz through it on a break in far less. While that seems steep, compared to a degree or bootcamp this micro-certification is a steal!
If you have a learning budget, or are dedicated to upskilling your career with a machine-learning focus– we recommend paying for and completing the program to get the shareable certificate (GET RECEIPTS!). This will help make your Linkedin more searchable to recruiters who may be looking for specific keywords and programs.
To audit the program and simply learn the material, this program is completely free! Thanks Coursera!
While this program was initially introduced to Stanford in 2012, it's only been a part of Coursera for a limited time (2022, to be exact!). However, the reviews are overwhelmingly positive-- which is to be expected from a tried-and-true program. Each of the three courses had a 4.9 rating average with thousands of reviews.
Some of our favorite positive review points:
- Perfect balance of application and theory, and wise choices in ramping up the complexity gradually. Discussion boards are very helpful, feels very much like personalized learning.
- Prof Ng is a fantastic teacher! The three courses are really well structured and builds upon themselves. I expected to learn some cool things, and I sure did - some mind-blowing machine learning things! The mentors on the forum are really helpful and respond to questions will thoughtful replies, which is great.
Prof Ng is passionate about machine learning, but is also sincere & humble, and is also very mindful of the ethics of AI and how it impacts people. The course is pretty cheap, and I can tell Prof Ng really wants to pass on AI knowledge.
Aggregations of negative review points:
- Assignments could be more difficult, and less google-able.
- Definitely at a pace for beginners.
... and our favorite overall review:
The best thing this course did for me was to remove the enigma of machine learning. This specialization is not so much about going deep into individual machine-learning algorithms and techniques as it is about exposing a student to the broad spectrum of all the different kinds of problems for which machines can be programmed to learn a solution.
Once a student completes this program, they have a very good idea of the kinds of problems that can be solved by letting machines learn how to solve those problems and specific algorithms/techniques that need to be used for that particular kind of problem. A student can then research additional resources for the specific problem they have at their hand and take a deep dive into developing a working solution for their specific problem. This course enables you to start that journey by taking away the fear created by the belief that machine learning is something very challenging.
This class is sponsored by U of M, and has a great rating of 4.8 with 1.5k reviews. With a Coursera subscription (included with your product management specialization, may we add!), you can get a certificate OR audit for free! Read our full writeup here.
UVA Sponsors a prestigious specialization program through Coursera. This program focuses on test and learn strategies, identifying and acting on customer insights, and running and effective product program. Read our full writeup here. It's free to audit, but if you want the certificate (recommended), it's $79 a month to complete at your own pace.
Other Analytics & ML Specializations
Google also sponsors a data analytics certificate program through Coursera. This is one of the more coveted certificates in the industry for learning the Google Analytics tool specifically, hence our recommendation of this course from Duke. Google's course is also free to audit, but same rules apply if you want the certificate to show off at $49 a month.
The University of Minnesota also runs a great Coursera program with a 4.7 star average. It's free to audit, but if you want the certificate it's covered under a $49/month Coursera subscription. We especially love course 2 for the experimenters out there... you can never go wrong with data-driven optimization strategies.
This is a newer beginner-level class that has a great overview of types of analytics, and when to use each method to maximize effectiveness.
Here at Bridged we are huge fans of stacking micro-certifications to achieve desired career results. By the end of this program, you'll be a machine learning master and have the practical skills you need to take on any real-world problem with confidence.If you're looking to break into the exciting field of AI or build a career in machine learning, this is the perfect place to start.
The new Machine Learning Specialization could be your ticket to a successful technical career.