REVIEW 3 cited by
Robust Algorithmic Collusion
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
This paper develops a formal framework to assess policies of learning algorithms in economic games. We investigate whether reinforcement-learning agents with collusive pricing policies can successfully extrapolate collusive behavior from training to the market. We find that in testing environments collusion consistently breaks down. Instead, we observe static Nash play. We then show that restricting algorithms' strategy space can make algorithmic collusion robust, because it limits overfitting to rival strategies. Our findings suggest that policy-makers should focus on firm behavior aimed at coordinating algorithm design in order to make collusive policies robust.
Forward citations
Cited by 3 Pith papers
-
Shared Bidding Algorithms and Competition: Evidence from Electricity Markets
Shared autobidding providers cause rival batteries to forgo profitable dispatch that would hurt same-provider rivals, with an estimated weight near one on rivals' profits above ~30% near-margin capacity share.
-
Algorithmic collusion under asynchronous price updating
Asynchrony in price updates hampers algorithmic collusion in simulated duopolies, except when algorithms monitor their competitor's current price.
-
Auditing Algorithmic Collusion from Strategy Graphs
Maximum betweenness and attractor in-degree of strategy graphs correlate with algorithmic collusion across Q-learning policies, offering a benchmark-free detection screen.
Discussion (0). Continue with ORCID to comment.