Machine learning educational content is often in the form of academic papers or blog articles. These resources are incredibly valuable. However, they can sometimes be lengthy and time-consuming. If you just want to learn basic concepts and don’t require all the math and theory behind them, concise machine learning videos may be a better option. The Youtube videos on this list cover concepts such as what machine learning is, the basics of natural language processing, how computer vision works, and machine learning in video games.
1. What Is Machine Learning? | Machine Learning Basics
Upload Date: September 19th, 2018
This video by Simplilearn provides a clear and simple explanation of many basic concepts of machine learning. In the video, Simplilearn explains the difference between supervised learning, unsupervised learning, and reinforcement learning. Additionally, they provide a brief explanation of the K-nearest neighbors algorithm. The video uses good visuals, graphs, and simple examples. Lastly, there is a quick quiz and a short overview of some of the most interesting applications of machine learning.
2. The Basics of Natural Language Processing (NLP)
Upload Date: June 6th, 2018
AI solutions provider SparkCognition explains the basics of NLP in this short video, with a runtime of under five minutes. The amount of information they are able to pack into such a short amount of time is admirable.
The video uses great visuals and animations to help create clear and concise explanations. It briefly touches on structured vs. unstructured data, stop words, and how NLP improves search engines. Lastly, SparkCognition explains how businesses can use NLP to analyze their data and boost operational efficiency and safety.
3. How Computer Vision Works
Upload Date: April 19th, 2018
Channel: Google Cloud Platform
Computer vision is one of the largest fields of research within machine learning. This video from Google provides a deep, yet concise, explanation of how computer vision works. In the video, they explain how computers see images and how machine learning models are trained to recognize objects.
Furthermore, there is a brief explanation of convolutional neural networks (CNN) and how they use labeled training data to make predictions and improve accuracy. Lastly, the video touches on how CNNs are good at understanding spatial features, but struggle with temporal features. With a runtime of just over seven minutes, Google manages to pack in a lot of information with great visual aids.
4. Multi-Agent Hide and Seek
Upload Date: September 17th, 2019
From OpenAI comes this incredibly interesting experiment on how multi-agent competition can lead to intelligent behavior.
In this video, OpenAI explains how they built a virtual world of hide and seek with simple rules and rudimentary tools. Within this world, they placed multiple agents with the role of either hider or seeker. After millions of rounds, the agents started to learn how to use tools to their advantage, through reinforcement learning. They began collaborating and finding creative ways to win the game. Eventually, they even learned how to use bugs in the programming to cheat their way to a victory. An alternate video with more commentary about the experiment can be found on the Two Minute Papers channel.
5. MarI/O – Machine Learning in Video Games
Upload Date: June 13th, 2015
In this video, Twitch streamer and computer programmer SethBling introduces MarI/O, an AI program he built that learned how to play Super Mario World.
With a demo of the program showing off its gaming skills, Seth explains neural network basics and MarI/O’s 24-hour evolutionary learning system. He goes over how the neural network was trained to play the game, what the neural network sees, and how it learned to make decisions and evolve over time.
6. What’s the Difference Between AI, Machine Learning, and Deep Learning
Upload Date: Aug. 10th, 2020
Channel: Machine Learning 101
If you’re looking for a quick primer on the exact definitions of AI, machine learning, and deep learning, tehn Machine Learning 101’s five minute video is a fun walkthrough of what they mean in both theory and practice. It’s full of easy to follow visual examples of practical AI applications, and the results in a short but comprehensive exploration of these popular AI terms.
7. Computer Vision and 5 Main Types of Image Annotation
Upload Date: Aug. 26th, 2020
Channel: Machine Learning 101
Image annotation is an integral part of the machine learning process for numerous computer vision models. This video will introduce five of the most common types of image annotation: bounding boxes, cuboids, polygons, lines, and semantic segmentation.
8. OpenAI’s New Language Generator: GPT-3
Upload Date: July 27th, 2020
Channel: What’s AI
One of the biggest developments in the Natural Language Processing sector was Open AI’s release of GPT-3. With 175 billion parameters, this text-generating model is incredibly powerful and is making waves online. Learn more about it from this video.
9. The State of AI in 2020
Upload Date: Oct. 1st, 2020
Channel: Machine Learning 101
In this podcast, Rahul Agarwal at WalmartLabs in India and Olaf Witkowski, the Director of Research at Cross Labs in Tokyo go over the current state of AI in 2020.
This podcast covers:
- How close are we to Artificial General Intelligence (AGI) or True AI?
- What problems are hindering AI research?
- What are the best recent advancements in machine learning?
- What needs to change before we can achieve AGI?
Under 10 minutes each, these machine learning Youtube videos are perfect to help pass the time during your commute or lunch break, while strengthening your knowledge of machine learning basics. After watching the videos above, you should have a better understanding of what machine learning is, the basics of NLP and computer vision, and machine learning in video games.
Looking for more articles on machine learning? See the 20 Best YouTube Channels for AI and Machine Learning.