Scale-Invariant Fast Convergence in Games

Scale-invariance in games has recently emerged as a widely valued desirable property. Yet, almost all fast convergence guarantees in learning in games require prior knowledge of the utility scale. To address this, we develop learning dynamics that achieve fast convergence while being both scale-free, requiring no prior information about utilities, and scale-invariant, remaining unchanged under positive rescaling of utilities. For two-player zero-sum games, we obtain scale-free and scale-invariant dynamics with external regret bounded by tildeO(Adiff), where Adiff is the payoff range, which implies an tildeO(Adiff / T) convergence rate to Nash equilibrium after T rounds. For multiplayer general-sum games with n players and m actions, we obtain scale-free and scale-invariant dynamics with swap regret bounded by O(Umax log T), where Umax is the range of the utilities, ignoring the dependence on the number of players and actions. This yields an O(Umax log T / T) convergence rate to correlated equilibrium. Our learning dynamics are based on optimistic follow-the-regularized-leader with an adaptive learning rate that incorporates the squared path length of the opponents' gradient vectors, together with a new stopping-time analysis that exploits negative terms in regret bounds without scale-dependent tuning. For general-sum games, scale-free learning is enabled also by a technique called doubling clipping, which clips observed gradients based on past observations.