In stylized competitive markets with noisy demand and iterated least squares learning, oblivious demand models yield transient collusive patterns that dissipate under sufficient exploration, informed sellers strictly outperform, and the modeling choice has a unique Nash equilibrium at the all-inform
arXiv preprint arXiv:2409.03956 , year=
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
verdicts
UNVERDICTED 6representative citing papers
Derives first lower bound on γ_t for mean-based algorithms in unknown-horizon bandit settings, proposes two new algorithms, and shows some are also no-regret.
No-swap-regret players frequently receive lower utilities than no-regret players in two-player games due to slower effective learning rates, though the reverse holds in some random 7-action games.
Bayesian learners can drive out no-regret learners despite logarithmic regret in stochastic markets, but no-regret is more robust; hybrids are proposed to combine strengths.
Domination-Avoiding agents provably avoid collusion in repeated price-competition markets and avoid playing strategies eliminated by iterated elimination of dominated strategies in any game.
Misspecified estimate-then-optimize pricing converges to supra-competitive prices when initial random explorations occur in similar ranges, reaching monopoly levels under symmetry.
citing papers explorer
-
Should Demand Models Incorporate Competitor Prices? Oblivious Learning and Algorithmic Collusion
In stylized competitive markets with noisy demand and iterated least squares learning, oblivious demand models yield transient collusive patterns that dissipate under sufficient exploration, informed sellers strictly outperform, and the modeling choice has a unique Nash equilibrium at the all-inform
-
Mean-based algorithms: A lower bound and regret
Derives first lower bound on γ_t for mean-based algorithms in unknown-horizon bandit settings, proposes two new algorithms, and shows some are also no-regret.
-
Hierarchies of No-regret Algorithms
No-swap-regret players frequently receive lower utilities than no-regret players in two-player games due to slower effective learning rates, though the reverse holds in some random 7-action games.
-
Markets with Heterogeneous Agents: Dynamics and Survival of Bayesian vs. No-Regret Learners
Bayesian learners can drive out no-regret learners despite logarithmic regret in stochastic markets, but no-regret is more robust; hybrids are proposed to combine strengths.
-
Domination-Avoiding Learning Agents Cannot Collude
Domination-Avoiding agents provably avoid collusion in repeated price-competition markets and avoid playing strategies eliminated by iterated elimination of dominated strategies in any game.
-
Misspecified Estimate-then-Optimize Leads to Supra-Competitive Prices
Misspecified estimate-then-optimize pricing converges to supra-competitive prices when initial random explorations occur in similar ranges, reaching monopoly levels under symmetry.