Prototype initial RL architecture (RL model)

[sdley]

Implement base model using attention, LSTM decoder, and pointer network as described in LIN202.

[dimeth-nouicer]

I suggest you start implementing a REINFORCE model instead. No need for attention or LSTM.

[sdley]

Well noted, Dimeth.
I'll take it in account.
Thanks