[nojvek]

What I would love to see is an end to end neural network soln. On one end camera input comes through, on the other outputs for speed and steering angle.

But rather than a black box, it's explainable what the different layers are doing. If neural nets are turing machines then we should be able to compile some parts of the net from code.

Then the net is a library of layers. Some Layers trained with back prop, some compiled from code.

[argonaut]

Almost all neural nets are not Turing complete. Only very specific RNNs are; most RNNs aren't, including pretty much any RNN model used in the real world right now (https://uclmr.github.io/nampi/talk_slides/grefenstette-nampi...).

Also, this is a useless fact, because so many other random things are Turing complete.

[wroscoe]

I'm working on adding the throttle. This is difficult because you need to drive the correct speed and stopping or running off course can mess up the training data. This project was inspired by Otavio's carputer which does predict throttle, steering angle, and odometer.

[deepnotderp]

The end to end approach of regressing steering wheel angle already exists, check nvidia's paper.