- 10.1 - Skipped
- 10.2 - Skipped
- 10.3 - Skipped
- 10.4 - Skipped
- 10.5 - Skipped
- 10.6 - Skipped
- 10.7 - Skipped
This is a comprably in-depth explanation of NN, my time is better spent doing labs or more chapters. I should come back and finish these but for now it is OK. Also I cannot be asked to do this right now.
Take home lessons:
- Artificial Neural networks and the tools associated with them have become a powerful tool to skip modeling a system using first principles, but simply learn its properties from data. as such they are capable of replacing many of the models discussed in previous chapters, ranging from kinematics to vision, feature detection, and controls.
- Simple neural networks are capable of both classification and regression akin to techniques described in Chapter 9, whereas convolutional networks are capable of filtering and pre-processing techniques such as described in Chapter 8.
- When a system is not purely reactive but requires state such as those described in Chapter 11, recurrent neural networks are needed to implement a notion of memory.