| Description |
-
I argue that the thrill of winning, which I call Endogenous Reinforcement, is adaptive in that it prevents agents from adopting pure strategies in competitive environments, which would leave them open to exploitation. Thus, Endogenous Reinforcement can facilitate Protean Behavior. This means that humans' learning and decision making apparatus is implicitly strategic, in a way that may generate ... read moresub-optimal behavior in the fact of pure randomness. What appears to be bounded rationality may therefore be strategic paranoia. This thesis shows three very different attempts I have made at modeling endogenous reinforcement, and its ability to help agents play mixed strategies, or at least avoid pure strategies.read less
|
|---|
This object is in collection
