Life of an AI project - MFML Part 2
Ғылым және технология
Making Friends with Machine Learning was an internal-only Google course specially created to inspire beginners and amuse experts. Today, it is available to everyone! This video is the second hour-and-a-half of a six hour session, covering the first half of "Lifecycle of an AI project" (Steps 0-5 of our 12 step applied AI guide).
The course is designed to give you the tools you need for effective participation in machine learning for solving business problems and for being a good citizen in an increasingly AI-fueled world. MFML is perfect for all humans; it focuses on conceptual understanding (rather than the mathematical and programming details) and guides you through the ideas that form the basis of successful approaches to machine learning. It has something for everyone!
Part 1 is available at bit.ly/mfml_part1
Part 3 is available at bit.ly/mfml_part3
To stay tuned for Part 4, don't forget to hit that that subscribe+notify button!
Looking for hands-on ML/AI tutorials? Here are some of my favorite 10 minute walkthroughs:
AutoML - console.cloud.google.com/?wal...
Vertex AI - bit.ly/kozvertex
AI notebooks - bit.ly/kozvertexnotebooks
ML for tabular data - bit.ly/kozvertextables
Text classification - bit.ly/kozvertextext
Image classification - bit.ly/kozverteximage
Video classification - bit.ly/kozvertexvideo
Пікірлер: 29
This is the most approachable, comprehensive machine learning content I've seen. The analogies are powerful, useful, and applicable. I also appreciated your emphasis on the important role played by cross-functional teams of human decision-makers and qualified 'Wish Makers.' The human element, I've found, is often overlooked and unappreciated. Thanks for sharing!
@xyx982
7 ай бұрын
5⁵
@xyx982
7 ай бұрын
⁵
Data science introduced as Ballet Dance! well done
Amazing content, the most simplified way of teaching I have encountered so far.
Cassie such a wonderful way to explain, I am clear about each and everything. Thank you.
You make it so intuitive and fun! Thank you.
Excellent content! Thanks for sharing!
It is true that the terminology can be really confusing whereas the actual thing it refers to could be simply explained even to a child. Thank you for the great material!
I love the content that you share on youtube. Thanks Csasie
Thanks for sharing this content Cassie, this is very inspiring to get to present ML to non tech persons :) Sharing it right now
Couldn't be better! Well articulated :)
her laptop stickers are dope af
This is really great content and I am truly inspired by it! Are any notes, blog post or slide deck available for quickly referencing the content for future use?
@dzivkovi
2 жыл бұрын
I p
Very well done. View count and comment count suggest that ML and AI is still too challenging for most.
Thank you !
SPLIT YOUR DAMN DATA!! Point taken lol. Loving the content.
Thanks Cassie, waiting for new content from you.
George Babbage was asked by an MP (funding his Difference Engine) "If the machine is given the wrong numbers, will it return the right answer?". By the sound of it, decision-makers haven't improved over a couple of centuries.
28:01 : Most imp first step 37:42 : Did it get it right or not?, did it found a cat? Precision vs recall is very important 1:21:04 : It is most important line for Applied ML. How long does each one take to try.
Where can the presentations be found ?
I like your voice, I can hear you all day and night
Cassie can we have the slides somehow? thank you!
Cat, or not cat, and the tiger. Now, think about how we get intuition about numbers. 5 digits of the hand, 3 the leaf clover, 2 wings of the bird...the 5 can be saw on 5 cats, the 3 in three units of cats, the 2 on two cats and so on...what I mean? About generalization...this power to generalize, acquire something that is a pattern among the things, we, humans, can do trough some examples...the machine can do from more examples (data)...what are the patterns that the AI had discovered and that we, humans, can not see? Some examples?
1:22:50
Not dog, not dog, not dog, not dog I feel dogs are underrepresented, this is systemic.
Programmers were simulating before staticians thought it was cool. ;)
Are we, humans, training AI with "No CAPTCHA reCAPTCHA"?