Upgrade to Pro — share decks privately, control downloads, hide ads and more …

A Gentle Introduction to Deep Learning for Deve...

A Gentle Introduction to Deep Learning for Developers at Abstractions

Like flying cars, the promise of artificial general intelligence (AGI) has been elusive for years. We are a long ways out from machines that can perform intellectual tasks equivalent to that of a human. Deep learning, a subset of machine learning, is a technology available today which enables developers to add narrow artificial intelligence (AI) capabilities to their applications including image or audio classification, facial recognition, object recognition, image caption generation, and natural language processing. We'll explore how developers can integrate pre-trained open source deep learning models into their applications and how developers and data scientists can collaborate on the development, training, and deployment of deep learning models.

Bradley Holt

August 21, 2019
Tweet

More Decks by Bradley Holt

Other Decks in Technology

Transcript

  1. A Gentle Introduction to Deep Learning for Developers — Bradley

    Holt Program Manager, Developer Advocacy Center for Open-Source Data & AI Technologies ↳ (CODAIT) @BradleyHolt | @ibmcodait | @IBMDeveloper medium.com/codait github.com/codait | github.com/IBM developer.ibm.com
  2. Is AI overhyped? Center for Open-Source Data & AI Technologies

    (CODAIT) / August 21, 2019 / © 2019 IBM Corporation
  3. Yes Center for Open-Source Data & AI Technologies (CODAIT) /

    August 21, 2019 / © 2019 IBM Corporation
  4. Thank you for coming to my talk Center for Open-Source

    Data & AI Technologies (CODAIT) / August 21, 2019 / © 2019 IBM Corporation
  5. ↳ General Artificial Intelligence Metal Skull With Terminator Eye by

    L.C. Nøttaasen, on Flickr <https://flic.kr/p/6xh2Dr> (CC BY-SA 2.0). Center for Open-Source Data & AI Technologies (CODAIT) / August 21, 2019 / © 2019 IBM Corporation
  6. ↳ Broad Artificial Intelligence -[ electrIc b88Gal88 ]- by JD

    Hancock, on Flickr <https://flic.kr/p/6x9t1H> (CC BY 2.0). Center for Open-Source Data & AI Technologies (CODAIT) / August 21, 2019 / © 2019 IBM Corporation
  7. ↳ Narrow Artificial Intelligence Danbo on the Lookout by IQRemix,

    on Flickr <https://flic.kr/p/x5oWjP> (CC BY-SA 2.0).
  8. Approaches to Artificial Intelligence Center for Open-Source Data & AI

    Technologies (CODAIT) / August 21, 2019 / © 2019 IBM Corporation Machine Learning • Predictive analytics • Data mining • Anomaly detection • Email filtering Deep Learning • Image or audio classification • Facial recognition • Object recognition • Image caption generation • Natural language processing
  9. Demo: ⬐ Object identification and image segmentation with magicat Center

    for Open-Source Data & AI Technologies (CODAIT) / August 21, 2019 / © 2019 IBM Corporation
  10. Install magicat $ npm install -g magicat + [email protected] added

    255 packages from 209 contributors in 11.798s
  11. Animal photos by Susanne Nilsson, on Flickr <https://www.flickr.com/photos/infomastern/> (CC BY-SA

    2.0). 23895621638_535be71dee_k.jpg 37038669284_899d7784a9_k.jpg 37489697170_31d05aa027_k.jpg 37699459356_24fd526a5e_k.jpg 37699976806_5ce694be36_k.jpg
  12. Scan Directory with magicat $ magicat . --contains sheep Scanning

    directory '~/tfjs-demos/magicat' for sheep... Sheep found in: 37038669284_899d7784a9_k.jpg 37489697170_31d05aa027_k.jpg
  13. Save Image Segment with magicat $ magicat 37699976806_5ce694be36_k.jpg --save horse

    The image '37699976806_5ce694be36_k.jpg' contains the following segments: background, horse. saved 37699976806_5ce694be36_k-horse.png Animal photos by Susanne Nilsson, on Flickr <https://www.flickr.com/photos/infomastern/> (CC BY-SA 2.0).
  14. Demo: ⬐ Veremin, A Video Theremin Based on PoseNet Center

    for Open-Source Data & AI Technologies (CODAIT) / August 21, 2019 / © 2019 IBM Corporation
  15. Software Programming Center for Open-Source Data & AI Technologies (CODAIT)

    / August 21, 2019 / © 2019 IBM Corporation Untitled by Marcin Wichary, on Flickr <https://flic.kr/p/68Lhc9> (CC BY 2.0).
  16. Center for Open-Source Data & AI Technologies (CODAIT) / August

    21, 2019 / © 2019 IBM Corporation 01011001001011000001 10101101010100100110 00010110101010011100 10101010110101010100 10001001011101000101 Software Program Business Logic
  17. Center for Open-Source Data & AI Technologies (CODAIT) / August

    21, 2019 / © 2019 IBM Corporation 01011001001011000001 10101101010100100110 00010110101010011100 10101010110101010100 10001001011101000101 Software Program Input Program Execution Output
  18. Machine Learning vs. Software Programming Center for Open-Source Data &

    AI Technologies (CODAIT) / August 21, 2019 / © 2019 IBM Corporation Machine Learning • Training • High-quality examples (i.e., training data) • Approximation of a correct function Software Programming • Programming • Set of direct instructions • Precisely-defined function
  19. Center for Open-Source Data & AI Technologies (CODAIT) / August

    21, 2019 / © 2019 IBM Corporation Input Layer Hidden Layers Output Layer
  20. Backpropagation Labeled Training Data Coat Sneaker T-shirt Sneaker Pullover Output

    Errors Pullover Coat Coat Sneaker T-shirt ❌ ❌ ❌ Fashion-MNIST dataset by Zalando Research, on GitHub <https://github.com/zalandoresearch/fashion-mnist> (MIT License).
  21. Input Output Sneaker 98% Neural Network Inferencing Fashion-MNIST dataset by

    Zalando Research, on GitHub <https://github.com/zalandoresearch/fashion-mnist> (MIT License).
  22. Machine Learning Libraries Center for Open-Source Data & AI Technologies

    (CODAIT) / August 21, 2019 / © 2019 IBM Corporation Untitled by Marcin Wichary, on Flickr <https://flic.kr/p/68Lhc9> (CC BY 2.0). The Leeds Library by Michael D Beckwith, on Flickr <https://flic.kr/p/r6ip1G> (CC0 1.0).
  23. Applying Deep Learning Center for Open-Source Data & AI Technologies

    (CODAIT) / August 21, 2019 / © 2019 IBM Corporation Untitled by Marcin Wichary, on Flickr <https://flic.kr/p/68Lhc9> (CC BY 2.0). Sharpest tool in the shed by Lachlan Donald, on Flickr <https://flic.kr/p/fkmB7T> (CC BY 2.0).
  24. Center for Open-Source Data & AI Technologies (CODAIT) / August

    21, 2019 / © 2019 IBM Corporation Data Science Expertise Computing Resources High-Quality Training Data Model Deployment Time Model Integration Inferencing Code And more… Sharpest tool in the shed by Lachlan Donald, on Flickr <https://flic.kr/p/fkmB7T> (CC BY 2.0).
  25. Center for Open-Source Data & AI Technologies (CODAIT) / August

    21, 2019 / © 2019 IBM Corporation Microservice Choose deployable model Deep Learning asset from the Model Asset Exchange (MAX) Deploy Swagger specification Inference endpoint Metadata endpoint Input preprocessing, model execution, and output post-processing Deploy model Use model https://developer.ibm.com/exchanges/models/
  26. AI Lifecycle Center for Open-Source Data & AI Technologies (CODAIT)

    / August 21, 2019 / © 2019 IBM Corporation Deploy & Run Operate & Manage Prepare, Build & Train ⇄ ↳ ⬏
  27. ⇄ Operating and Managing AI Systems Fabric for Deep Learning

    (FfDL) Train and deploy deep learning models on Kubernetes using TensorFlow, Caffe2, PyTorch, and other frameworks Trusted AI • AI Fairness 360 Toolkit: Fairness metrics for machine learning models, explanations for these metrics, and algorithms to mitigate bias • Adversarial Robustness 360 Toolbox: Python library for adversarial attacks and defenses for neural networks • AI Explainability 360 Toolkit: Interpretability and explainability of data and machine learning models
  28. What will you build with deep learning? Dynamic Earth -

    Continental Shelf by NASA Goddard Space Flight Center, on Flickr <https://flic.kr/p/ch8t25> (CC BY 2.0).
  29. Resources: Machine Learning Libraries TensorFlow https://www.tensorflow.org/ Train your first neural

    network: basic classification | TensorFlow https://www.tensorflow.org/tutorials/keras/basic_classification Keras https://keras.io/ PyTorch https://pytorch.org/ Caffe2 https://caffe2.ai/
  30. Resources: IBM Developer IBM Developer https://developer.ibm.com/ IBM Cloud https://ibm.biz/BdzXfW Center

    for Open-Source Data & AI Technologies (CODAIT) http://codait.org/ IBM Developer Model Asset Exchange (MAX) https://developer.ibm.com/exchanges/models/ Model Asset Exchange (MAX) Code Patterns https://developer.ibm.com/patterns/category/model-asset-exchange/ magicat https://github.com/CODAIT/magicat Veremin https://veremin.mybluemix.net/
  31. Resources: Preparing, Building, and Training AI Models Project Jupyter https://jupyter.org/

    IBM Developer Model Asset Exchange (MAX) https://developer.ibm.com/exchanges/models/ IBM Watson Knowledge Catalog https://www.ibm.com/cloud/watson-knowledge-catalog IBM Watson Studio https://www.ibm.com/cloud/watson-studio
  32. Resources: Deploying and Running AI Models Data Mining Group (PMML

    & PFA) http://dmg.org/ ONNX https://onnx.ai/ ONNX.js https://github.com/Microsoft/onnxjs TensorFlow.js https://js.tensorflow.org/ TensorFlow Lite https://www.tensorflow.org/lite Core ML https://developer.apple.com/machine-learning/ IBM Watson Machine Learning https://www.ibm.com/cloud/machine-learning
  33. Resources: Operating and Managing AI Systems Fabric for Deep Learning

    (FfDL) https://github.com/IBM/FfDL AI Fairness 360 Toolkit https://github.com/IBM/AIF360 Adversarial Robustness 360 Toolbox https://github.com/IBM/adversarial-robustness-toolbox AI Explainability 360 Toolkit https://github.com/IBM/AIX360/ IBM Watson OpenScale https://www.ibm.com/cloud/watson-openscale
  34. Thank you. Center for Open-Source Data & AI Technologies (CODAIT)

    / August 21, 2019 / © 2019 IBM Corporation Bradley Holt Program Manager, Developer Advocacy Center for Open-Source Data & AI Technologies ↳ (CODAIT) — @BradleyHolt | @ibmcodait | @IBMDeveloper medium.com/codait github.com/codait | github.com/IBM developer.ibm.com