A two-stage feature-based belief aggregation compresses POMDP beliefs and gives a cost approximation whose error is at most epsilon divided by (1 minus the discount factor).
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SY 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Feature-Based Belief Aggregation for Partially Observable Markov Decision Problems
A two-stage feature-based belief aggregation compresses POMDP beliefs and gives a cost approximation whose error is at most epsilon divided by (1 minus the discount factor).