Deep Learning with Actuarial Applications in R

Course material of the block course "Deep Learning with Actuarial Applications in R" of the Swiss Association of Actuaries, held in 2020 and 2021 in Zurich. The course material consists of:

• Slides for the theoretical part (all in one pdf)
• Complete code of the exercises as .R,.html,.Rmd, .ipynb files
• Data used in the exercises (data)
• Introduction to keras (keras)

Topics:

• Topic 1: Generalized Linear Models (glm)
• Topic 2: Feed-forward Neural Networks (fnn)
• Topic 3: Discrimination-Free Insurance Pricing (dfp)
• Topic 4: LocalGLMnet (lgn)
• Topic 5: Convolutional Neural Networks (cnn)
• Topic 6: Recurrent Neural Networks (rnn)

Download:

To download all the material in .zip file, press the green button "Code" at the top right of this page and select "Download ZIP".

References:

Please find below links for supporting literature:

• "Actuarial Data Science", Swiss Association of Actuaries, https://www.actuarialdatascience.org
• "Keras", RStudio, https://keras.rstudio.com/
• "Deep Learning with Keras :: CHEAT SHEET", RStudio, https://github.com/rstudio/cheatsheets/raw/master/keras.pdf