AI for Beginners from a Beginner - PowerPoint Presentation

AI for Beginners from a Beginner

About me AI beginner tomas@florian.caIT Consulting Complex networkingCloud / Virtualization systems Cyber security

Demos What got me started down this path is impressive demos that I’ve seen in the last couple of years

Question Can anybody do this with now with open source software? Yes

Lay of the Land Closed Cloud Open Source Research Libraries Frameworks CLI apps End user apps

Lay of the Land Closed Cloud Open Source Research Libraries Frameworks CLI apps End user apps

Lay of the Land Closed Cloud Open Source Research Libraries Frameworks CLI apps End user apps

Lay of the Land Closed Cloud Open Source Research Libraries Frameworks CLI apps End user apps

Lay of the Land Closed Cloud Open Source Research Libraries Frameworks CLI apps End user apps

2x Demo Show UnwrapHowtoQuestionsNavigating Limits of AI Questions

Demo #1 Object Detection

Object Detection

Object Detection

Unwrap: Black Box ./flow -- imgdir sample_img / --model cfg /yolo.cfg --load bin/ yolo.weights

Neural Network ./flow -- imgdir sample_img / --model cfg /yolo.cfg --load bin/ yolo.weights

Neural Network Model ./flow -- imgdir sample_img / --model cfg /yolo.cfg --load bin/ yolo.weights Weights (pre-trained)

Darkflow Dependency Stack Anaconda TensorFlow OpenCV Python3 Ubuntu 18.04 VM i7 CPU, 4 GB RAM darkflow model weights

Howto git clone https://github.com/thtrieu/darkflowCreate conda virtual env for the project conda create -n NAME python=3.6source activate NAMEInstall dependencies conda install tensorflow cython numpyAdd the repo with particular opencv versionconda config --add channels conda-forgeInstall opencvconda install opencv Run setup python3 setup.py build_ext -- inplace Download weight file for the model https://drive.google.com/drive/folders/0B1tW_VtY7onidEwyQ2FtQVplWEU and place it in bin/ Run ./flow -- imgdir sample_img / --model cfg /yolo.cfg --load bin/ yolo.weights

Questions?

Demo #2 Generative Adversarial Network (GAN)

ThisPersonDoesNotExist.com

Demo #2

Unwrap: Black Box

GAN Generative Adversarial Network Pre trained generator network

Transparent Latent GAN Anaconda Python3 Ubuntu 16.04 VM 2 CPU,6 GB RAM, K80 GPU with 12GB RAM,50 GB Disk Transparent latent GAN CUDA toolkit cuDNN Jupyter Pretrained network model

Howto Git clone https://github.com/SummitKwan/transparent_latent_gan.gitPrepare anacondaconda create -n NAME python=3.6source activate NAME cd transparent_latent_ganInstall dependenciesconda install pippip install -r requirements.txt conda install cudatoolkitconda install cudnnconda install jupyterDownload pre-trained model https://www.dropbox.com/sh/y1ryg8iq1erfcsr/AAB--PO5qAapwp8ILcgxE2I6a?dl=0 (extract to same folder structure) Run notebook jupyter notebookNavigate to URL shown at startup + notebooks/transparent_latent_gan/ src /notebooks/ tl_gan_ipywidgets_gui.ipynb Run notebook

Hardware Notes Google Compute Engine preemptive K80 ~$0.20 CAD / Hour I used vanilla Ubuntu 16.04 and installed Nvidia drivers on it myself Prebuilt images but more $ per hour (not preemptive)REMEMBER TO TURN IT OFF Nvidia GX 1060 (6 GB RAM … more is better) $300Nvidia-smi

Expectations 80% of time dealing with building the stack Poor documentationMissing/incompatible pre-trained modelsDependency hell (much better with Anaconda)Unhelpful error messages 20% real AI work

Path of least resistance Anaconda cuDNN 7CUDA toolkit 9Ubuntu 16.04nVIDIA GPU > 6 GB RAM

You can do this now

Questions

Limits Too good to be true?

CSI zoom and enhance

CSI Zoom and Enhance for real https://github.com/alexjc/neural-enhance

Zoom and Enhance

…Again

… Again

We got him – URL 937

We look up the registered owner Cops go out Shoot the guyCSI Calgary saves the dayCase closed

Let’s run the same image through a different model

Zoom and Enhance

What?

Wait … what?

Side by side

Different model will CONVINCINGLY lead you to a different conclusion

Which license plate was it? Maybe URL 937 Maybe BBL 3698Maybe SOMETHIN ELSE Even though we are seeing it in front of our own eyes there is a threshold at which AI can just make stuff up and make it look like the real thing

NN has learned to make convincing fakes Those fakes may be rooted in reality or they may be purely hallucinated into existence The degree to which it’s rooted in reality depends on the data it’s been exposed to during trainingGarbage in – elaborate bullshit out

In the news: Woody Herrelson Look alike The image of the suspect, taken from the surveillance footage, was extremely pixelated and turned up no results in Facial ID system Investigators used high-quality images of Harrelson found on Google and submitted them in place of the suspect's more pixelated image. The result ended in a match.An unidentified man was booked and charged for petty larceny.

So when is the AI apocolypse coming?

Risk of AI Create AI that succeeds at producing output that is apparently super-intelligent, super accurate, and super trustworthyWhen in reality the output is elaborately disguised bullshit Sells well but the world doesn’t need more of it

GAN Limit Demo

How GAN works

How GAN works

Be aware of AI limits and use it within those limits Use AI within it’s limits

Best Open Source AI in 2019 https://medium.mybridge.co/amazing-machine-learning-open-source-tools-projects-of-the-year-v-2019-95d772e4e985

Questions? Presentation copy: lab.florian.ca tomas@florian.ca

Useful Links AI Cheat sheet https://becominghuman.ai/cheat-sheets-for-ai-neural-networks-machine-learning-deep-learning-big-data-678c51b4b463

Other cool stuff Video Game graphics GAN https://thenewstack.io/deep-learning-ai-generates-realistic-game-graphics-by-learning-from-videos/ Style GAN https://www.youtube.com/watch?v=-R9bJGNHltQ