An explore-then-commit algorithm for submodular maximization under full-bandit feedback.
Guanyu NieMridul AgarwalAbhishek Kumar UmrawalVaneet AggarwalChristopher John QuinnPublished in: UAI (2022)
Keyphrases
- simulated annealing
- dynamic programming
- cost function
- objective function
- segmentation algorithm
- high accuracy
- worst case
- experimental evaluation
- learning algorithm
- search space
- preprocessing
- matching algorithm
- detection algorithm
- improved algorithm
- k means
- computational cost
- optimal solution
- theoretical analysis
- markov chain
- times faster
- classification algorithm
- expectation maximization
- significant improvement
- particle swarm optimization
- computational complexity
- np hard