Einführung in Machine Learning

Wieso Sie sich jetzt mit Machine Learning beschäftigen sollten


The Architecture Gathering 2019, Oktober 2019


Oliver Zeigermann / @DJCordhose

embarc Software Consulting GmbH

Slides: http://bit.ly/ml-tag

Programmer's approach: Code Rules by Hand


if age < 25:
    if speed > 140:
        return red # young people, fast cars: high risk
    else:
        return yellow # young people: medium risk
            

if age > 75:
    return red # old people: high risk
            

if miles_per_year > 30:
    return red # a lot of driving: high risk
if miles_per_year > 20:
    return yellow #  a bit of driving: medium risk
            

return green # otherwise: low risk

Let's check with existing data

Approaching with data




  • 0 - red: many accidents
  • 1 - green: few or no accidents
  • 2 - yellow: in the middle

Prediction with Decision Trees

Up to 70% accuracy on unknown data

Architectures of Convolutional Neural Networks: VGG

There are a number of specialized neural network layers

MNIST - Using a model already trained

Exploring the different types layers together

https://transcranial.github.io/keras-js/#/mnist-cnn

Thank you

The Architecture Gathering 2019: Wieso du dich jetzt mit Machine Learning beschäftigen solltest

Oliver Zeigermann / @DJCordhose
embarc Software Consulting GmbH
http://bit.ly/ml-tag