When I use Neural Automated Model Discovery for Autocompleting Models with Prior Structural Information from JuliaSimModelOptimiser.
How is the neural network connected with the incomplete model? Is it:
\frac{d states}{dt} = model_{incomplete}(states) + model_{network}(states)
Can I control this connection through JuliaSimModelOptimizer? Or do I always have to provide a network that uses the same number of inputs, outputs and states?
Currently the way the network is connected is not very flexible (it’s just added to all states as you mentioned), but we are looking to improve that.