Here are three great Machine Learning and Artificial Intelligence courses, two of them from prestigious academic institutions and one from Google, all available as free videos.

"Thanks" to the pandemic, many noteworthy college classes that would be only accessible to their registered institutional students are making their way to the public and for free.

We have already looked at Stanford's Natural Language Processing with Deep Learning and before that at Free Resources For Machine Learning  which included "Machine Learning with Graphs" from Stanford and "Introduction to Deep Learning" by Sebastian Raschka. Another three courses that we've examined before are:

Cornell's CS 6120 Advanced Compilers

Yann LeCun’s Deep Learning Course Free From NYU

Nottingham's University Functional Programming in Haskell

This time we are coming back with a few more.

First stop is Prof. Dr. Stefan Harmeling's of Heinrich-Heine-Universität Düsseldorf "Machine Learning Lecture WS 2021/22", WS for "Wintersemester". Despite being taught at a German University, no worries, the lectures are all in English.

Dr. Stefan Harmeling posts his lectures on YouTube as they  take place with the latest upload happening a couple of days ago. The accumulated lectures comprise a playlist of 15 at the time of writing:

01 - Introduction
02 - Plausible Reasoning, Bayes rule
03 - From Logic to Probabilities, Bayesian Networks
04 - d-Separation, Continuous Probabilities
05 - The Gaussian distribution
06 - Distributions, MAP, ML,
07 - Transformation of Variables, Linear Regression
08 - Matrix Differential Calculus
09 - Model selection - Training, Evaluation and Test sets
10 - SVM part 1 (separable case)
11 - SVM part 2 (non-separable case)
12 - SVM part 3 (nonlinear case)
13 - PCA
14 - kernel PCA
15 - ISOMAP and LLE

While the videos are available, the complementary resources like pdf's, power-points and lecture notes are not.

Next stop is "Computer Science 3200 - Fall 2021 - Intro to Artificial Intelligence" by Professor David Churchill of Memorial University.

Professor David Churchill goes one step further, livestreaming the lectures on Twitch! then making the recordings available on YouTube.

This course is an introduction to Artificial Intelligence (AI), covering algorithmic techniques and data structures used in modern problem-solving environments and comes with the following syllabus :

Introduction to Artificial Intelligence
o What is AI? What can Modern AI do?
o Games as a Testing Environment for AI
o Agents, Environments, and Problems

• Search Algorithms
o Exhaustive Search (BFS / DFS)
o Heuristic Functions / Incorporating Knowledge
o Heuristic Search (Best-First Search / A*)
o Introduction to Game Theory / Nash Equilibrium
o Adversarial Search (Minimax / Alpha-Beta)
o Data Structures / Optimizations for Search

• Genetic Algorithms (GA)
o Introduction to Evolutionary Algorithms
o GA Representations: (Genotype, Phenotype)
o GA Implementation: Mutation, Crossover, Selection, Reproduction

• Reinforcement Learning (RL)
o Introduction to RL: Agent, Environment, Actions, Policies, Rewards
o Bandit Problems (Exploration vs. Exploitation)
o Markov Decision Processes
o Generalized Policy Iteration
o Monte-Carlo Methods
o Temporal Difference Learning (SARSA / Q-Learning)

• Neural Networks (NN)
o Artificial Neurons / NN Structure / Training
o Brief Introduction to Deep Learning

Fortunately in this case all the complementary resources come attached and cleanly indexed in a shared Google spreadsheet.

Last, but certainly not least, is "Making Friends with Machine Learning". This is a previously internal-only Google course specially created to inspire beginners and amuse experts and is now available to everyone for free.

It is run by Cassie Kozyrkov, nothing short of a genius, who has four degrees; psychology,economics, mathematical statistics and cognitive neuroscience. Her role at Google is that of Head of Decision Intelligence and what she actually does is looking at data using the lenses of statistics and machine learning to help people and companies make informed decisions. If you watch the videos you'll discover that she has got a unique sense of humor too!

MFML is more down to earth and focuses on the conceptual understanding rather than the mathematical and programming details, guiding you through the ideas that form the basis of successful approaches to machine learning.

It's a six hour session that has been split into 4 installments:

• Introduction to ML and AI - MFML Part 1
• Life of an AI project - MFML Part 2
• Taking AI from prototype to production - MFML Part 3
• Guide to AI algorithms - MFML Part 4

The first three parts are an attempt to introduce the concepts in an easy to understand language while the fouth and final one is more technical as it introduces you to the inner workings of popular ML/AI algorithms:

• Clustering and k-Means
• Lazy learning and k-NN
• Perceptron
• Maximal Margin Classifier
• Support Vector Classifier
• Support Vector Machines
• Decision Trees
• Boosted Aggregation
• Random Forests
• Ensemble Models
• Naive Bayes
• Linear Regression
• Logistic Regression
• Neural Networks / Deep Learning

Cassie's playlist spans beyond those four lengthy videos and includes another 52 short clips on everything ML.

So there we have our treat - another three great resources to start your ML journey.

More Information

Machine Learning Lecture WS 2021/22

COMP3200 - Intro to Artificial Intelligence

Making Friends with Machine Learning

Related Articles

Stanford's Natural Language Processing with Deep Learning

Free Resources For Machine Learning

Cornell's CS 6120 Advanced Compilers

Yann LeCun’s Deep Learning Course Free From NYU

Nottingham's University Functional Programming in Haskell

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info