Ignite Talk: Machine Learning Exposed!
Download as PPTX, PDF1 like398 views
This document provides an overview of machine learning concepts including supervised learning, unsupervised learning, and reinforcement learning. It discusses examples like self-driving cars, image description generation, and the game Go. The document also demonstrates supervised learning classification using the Iris data set, showing the neural network training process of forward and back propagation, and using the trained network to make predictions. It concludes with thanks for attention to the topic of machine learning.
Ignite Talk: Machine Learning Exposed!
- 3. Machine Learning Exposed! (ignited?) James Weaver Developer Advocate – Pivotal @JavaFXpert
- 5. Generating image descriptions Credit: Andrej Karpathy, Li Fei-Fei http://cs.stanford.edu/people/karpathy/deepimagesent/
- 7. Supervised learning regression problem (from Andrew Ng’s Machine Learning course)
- 9. Unsupervised learning finds structure in unlabeled data (e.g. market segment discovery, and social network analysis)
- 10. Reinforcement
- 11. AlphaGo is a recent reinforcement learning success story Source: https://gogameguru.com/i/2016/03/AlphaGo-Lee-Sedol-game-3-game-over.jpg
- 12. Supervised learning classification problem (using the Iris flower data set) Sepal length Sepal width Petal length Petal width Species 5.1 3.5 1.4 0.2 Iris setosa 4.9 3.0 1.4 0.2 Iris setosa 7.0 3.2 4.7 1.4 Iris versicolor 6.4 3.2 4.5 1.5 Iris versicolor 6.3 3.3 6.0 2.5 Iris virginica 5.8 3.3 6.0 2.5 Iris virginica Features Labels Training / test data
- 13. Iris data classified in four dimensions Sepal length Sepal width Petal length Petal width Species 5.1 3.5 1.4 0.2 Iris setosa 4.9 3.0 1.4 0.2 Iris setosa 7.0 3.2 4.7 1.4 Iris versicolor 6.4 3.2 4.5 1.5 Iris versicolor 6.3 3.3 6.0 2.5 Iris virginica 5.8 3.3 6.0 2.5 Iris virginica Credit: Tal Galili https://cran.r-project.org/web/packages/dendextend/vignettes/Cluster_Analysis.html Features Labels Training / test data
- 14. Modeling the brain works well with machine learning (ya think?)
- 15. Neural network trained for Iris classification forward propagation
- 16. Neural network trained for Iris classification back propagation (minimize cost function)
- 17. Neural net visualization app (uses Spring and DL4J) https://github.com/JavaFXpert/visual-neural-net-server
- 18. Entering feature values for prediction (classification)
- 19. Feedforward process for prediction -normalize inputs
- 20. Feedforward process for prediction -weights and bias
- 21. Feedforward process for prediction -node activations
- 22. Machine Learning Exposed! (and thanks for your kind attention) James Weaver Developer Advocate – Pivotal @JavaFXpert