Simple actor-critic learning

[tcstewar]

Here's a nice actor-critic reinforcement learning model that would be fun to re-implement in Nengo (and try different learning rules)

https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1003024

Here's another somewhat similar RL model that'd be a good target too

https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1000586

[ndey96]

Gave this a shot here: https://github.com/ndey96/spiking-actor-critic