# Lecture 8 ## Topic In this lecture you learn what a neural network actually is and implement one from scratch. ## Lecture Slides ```{raw} html ``` ```{eval-rst} :download:`Download the slides <_static/bld/pdfs/lecture_8.pdf>` ``` ## Exercises ```{toctree} --- maxdepth: 1 --- bld/notebooks/exercises/exercise_8.ipynb bld/notebooks/solutions/exercise_8.ipynb ``` ## Suggested Homework - Watch the Video below - Play around with a different model (e.g. more layers, larger layers) - How small can you make the model before it breaks down? - Does a larger model help? - Play around with the hyperparameters of the optimization ## Additional materials ### 3Blue1Brown Video This excellent Video uses the exact same network you train in the exercises and gives explains it from a slightly different perspective. You should absolutely watch it, ideally before the lecture.