In performative Markov potential games, independent policy gradient and natural policy gradient algorithms are claimed to converge to an approximate performatively stable equilibrium, but the proofs contain a false equality between value differences and potential differences.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Independent Learning in Performative Markov Potential Games
In performative Markov potential games, independent policy gradient and natural policy gradient algorithms are claimed to converge to an approximate performatively stable equilibrium, but the proofs contain a false equality between value differences and potential differences.