Research Article

Limiting Dynamics for Q-Learning with Memory One in Symmetric Two-Player, Two-Action Games

Table 5

All strategy pairs solving the Bellman equations self-consistently for the hawk-dove game, their structural and behavioural types, the conditions for their existence, and the regions of Figure 7 in which they exist.

Policy-pairTypeConditionsRegions

Sym, CC
Com, NCCAlways
Com, NCCAlways
Com, NCC
Com, NCC
Com, NCC
Com, NCC
Com, NCC
Com, NCC

CS, NCC
CS, NCC
CS, NCC
CS, NCC
CS, NCC
CS, NCC
CS, NCC
CS, NCC