Optimal strategies maximizing utilitarian social welfare in MDPs with heterogeneous discount factors are pure finite-memory counting strategies synthesizable in polynomial time, while threshold questions for positional strategies are NP-hard.
2.Long-term values.The value functions underπ ∞ from the initial states 0 are: V0(s0) = 1072.2294andV 1(s0) = 1.00001
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.GT 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Social Welfare under Heterogeneous Time Preferences
Optimal strategies maximizing utilitarian social welfare in MDPs with heterogeneous discount factors are pure finite-memory counting strategies synthesizable in polynomial time, while threshold questions for positional strategies are NP-hard.