Ready to demystify the world of AI? In this video, we dive
into the practical aspects of building your very own AI model to detect
military tanks. We kick things off by explaining the basics of how AI 'sees'
and why convolutional neural networks (CNNs) are the go-to for image
recognition. You'll get an up-close look at the RGB pixels that make up an
image and how they transform into data points that a machine can understand.
Next, we introduce you to Jeremy Howard's revolutionary course, 'Practical Deep
Learning for Coders,' and guide you through setting up Jupyter Notebooks on
Kaggle. Using the FastAI library, we'll fine-tune a pre-trained ResNet18 CNN
model with just 200 images! The best part? All of this can be done for free and
in a matter of minutes. This video aims to empower acquisition professionals
with the know-how to better manage, acquire, and understand AI-driven systems.
Say goodbye to the notion that AI is only for the tech elites; it's time to
make AI practical for everyone.
If you would like to watch the "How AI Learns to See" videos, those can be found here: https://media.dau.edu/media/t/1_xvztjh10 and here: https://media.dau.edu/media/t/1_4k58zfye
The link to a course playlist
of DAU recommended AI courses is: https://dau.csod.com/ui/lms-learner-playlist/PlaylistDetails?playlistId=00118adb-20e1-4dc5-95a8-9ffd03ab7f70
You will need a DAU account to access these
resources. If you are a DoD member and need a DAU account you can request one
here: https://www.dau.edu/faq/p/New-DAU-Account
Referenced course, "Practical Deep Learning for Coders", can be
found here: https://course.fast.ai/