11 data science videos every data scientist should know

Share on facebook
Share on twitter
Share on linkedin
Share on telegram
Share on whatsapp

Contents

Overview

  • Presenting 11 data science videos that will enhance and expand your current skill set
  • We have categorized these videos into three fields: natural language processing (PNL), generative models and reinforcement learning.
  • Learn how the concepts in these videos work and create your own data science project!!

Introduction

I love to learn and understand Data science concepts through videos. I just don't have time to read books and pages of text to understand different ideas and topics. However, I get a much better overview of the concepts through videos and then select the topics I want to learn more about.

The great quality and diversity of themes available on platforms like YouTube never ceases to amaze. I recently learned about the amazing XLNet framework for NLP from a video (that I mentioned below for your consumption). This helped me understand the concept so that I could explore more about XLNet!!

Data Science Videos

I firmly believe that structure is very necessary when we are learning any concept or topic. Also I follow that approach every time I write a post. That is why I have categorized these videos in their respective domains, mainly Natural Language Processing (PNL), Generative Models and Reinforced Learning.

Then, Are you ready to dive in and explore the breadth and breadth of data science through these fascinating videos??

Without any more preambles, here there is 11 amazing data science videos:

  • Natural language processing (NLP)
    • XLNet explained
    • How does Google Duplex work?
    • POEMPORTRAITS from Google: combination of art and artificial intelligence
  • Generative models
  • Reinforced learning
    • Teach the computer to drive
    • Find out how AlphaGo Zero works
    • Google DeepMind AI learns to walk
    • AI learns to play 2048
  • FIRST
    • Adobe develops AI to detect faces with Photoshop

XLNet explained

XLNet is the most popular framework in NLP right now. You simply should consider what it is and how it works if you want to build a career in this field. I came across this video recently and wanted to share it with the community ASAP.

XLNet is the next generation NLP framework. It has surpassed Google's BERT in 20 NLP tasks and achieved cutting-edge results in 18 of them. That is very, very impressive.

Be sure to check out our post on XLNet and its powerful capabilities here.

The video below provides a clear explanation of the original XLNet research post. Note: You may need to know some NLP concepts beforehand to truly understand the inner workings of XLNet.

How does Google Duplex work?

Remember when Sundar Pichai took the stage and put the whole world in a frenzy when he introduced Google Duplex in his keynote address on Google search engine I / O 2018? I remember listening with total amazement to the super realistic calls that the AI ​​made.

It took a while for the data science and NLP community to find an explanation of how Google Duplex really works.. It is quite powerful and has the potential to change the way we interact with machines..

Then, the million dollar question: Did Google Duplex pass the Turing test? You decide after watching this video:

POEMPORTRAITS from Google: combination of art and artificial intelligence

I am an artist and the prospect of combining any art form with Artificial Intelligence is extremely attractive. In a world where there is so much fear around AI, these applications are more than welcome.

POEMPORTRAITS AI by Google has been trained in 19th century poetry using NLP techniques. You can contribute and donate a word to generate your own POEMPORTRAIT. See how this incredible concept works:

Generative models

Dive into Variational Automatic Encoders!

Here's one of our favorite reinforcement learning experts, Xander Streenbrugge, from his wonderful ArxivInsights channel.

Variational Autoencoders (Alas) are powerful generative models with various applications. You can generate human faces or synthesize your own music or use VAE to erase noise from images.

I really like this video. Xander begins with an introduction to basic automatic encoders and then moves on to unraveled VAE and beta-VAE. quite technical, but beautifully and concisely explained, in typical Xander style.

Xander will return to DataHack Summit this year so you can hear and meet him in person.

Create facial animation from audio

I was immediately drawn to the video when I read the title. This is the best of generative models!! You can not only generate facial animation from audio, but it can also generate different emotions for the same audio. And facial expressions look incredibly natural.

If you are not following Two Minute Papers, is missing it. Regularly, produce videos that analyze the latest developments in an easy to understand way. It's a gem of a channel.

MuseNet learned how to compose Mozart, Bon Jovi and more

Another entry from the Two Minute Papers file.

OpenAI's MuseNet is a deep neural network that generates musical compositions with different instruments and combines different styles. Uses the same general purpose unsupervised technology as GPT-2 and the results are amazing.

Have you never heard of GPT-2? It is an NLP framework on par with XLNet. See how MustNet works here:

Reinforced learning

Teach how to drive a computer

Self-driving cars have always fascinated me. The scale of the autonomous vehicle project is staggering. There are so many components, both on the hardware side and on the data science side, that need to be aligned for this project to work.

This is a perfect video for beginners to learn about genetic programming and reinforcement learning and how they are used to create powerful applications. Simon's personality kept me hooked until the end.

And I am definitely testing the project on my own.

Find out how Google DeepMind's AlphaGo Zero works

Another great video from Xander. Explains the popular Google DeepMind post about AlphaGo Zero.

AlphaGo Zero is a remake of the original AlphaGo program that beat the human champion Lee Sedol comprehensively.. I recommend reading our post on Monte Carlo Tree Search, the algorithm behind AlphaGo before proceeding to learn about AlphaGo Zero.

AlphaGo Zero uses Reinforcement Learning to beat the world's best Go players without using human game data.

“AlphaGo Zero surpassed AlphaGo Lee's strength in three days by winning 100 games to 0, reached the level of AlphaGo Master in 21 days and outperformed all old versions in 40 days”.

Source: Wikipedia

Google DeepMind AI learns to walk

This video is fun and informative. Exactly the kind of video I like when I'm learning new things!! It was fun watching the AI ​​learn to walk. But at the same time, I was in awe of the power of reinforcement learning.

The video analyzes 3 posts to try and explain how AI learned to walk and is surprisingly simple to understand.

AI learns to play 2048

Have you ever played the game 2048? It's super addictive once you get the hang of it. I used to easily finish games before, but not anymore. Being a data science enthusiast, i am going to train my computer to play it with the help of this amazing video.

This is another example of the use of genetic programming and evolutionary algorithms..

BONUS: Adobe develops AI to detect faces with Photoshop

Adobe is the market leader in image and video manipulation software. Other companies have tried, but not many have come close to Adobe's level.

Last month, Adobe announced its investigative efforts to detect tampered images. It was about time someone did that!! Soon it will be impossible to tell the real from the fake, given how quickly GANs have taken over the world.

Imagine Donald Trump challenging Kim Jong Un to nuclear war and then claiming it was a sham and ignoring all responsibility!! We need to prevent those situations from becoming reality. This video shows how Adobe's algorithm works and tried to combat fake images:

Final notes

I love learning about the latest research in data science, machine learning and artificial intelligence. But I find it hard to read the posts. It takes a lot of time and effort, something not all data science professionals have. I'm sure many of you struggle with the same. Consuming videos is the ideal way to get an overview of these concepts..

Then, you can select where your interests lie and try to develop a project or blog post about it. Créame, is a wonderful way to learn and ingrain new data science concepts.

What are some of your favorite channels or videos on data science?? Let's discuss in the comments below.

Subscribe to our Newsletter

We will not send you SPAM mail. We hate it as much as you.