Restart Policies with Dependence among Runs: A Dynamic Programming Approach.
Yongshao RuanEric HorvitzHenry A. KautzPublished in: CP (2002)
Keyphrases
- dynamic programming
- optimal policy
- partially observable markov decision processes
- random walk
- state space
- infinite horizon
- decision problems
- multistage
- markov decision problems
- markov decision process
- dynamic programming algorithms
- reinforcement learning
- markov decision processes
- stereo matching
- coarse to fine
- policy search
- single machine
- long run
- revenue management
- management policies
- uc berkeley
- optimal control
- sequence alignment
- lagrangian relaxation
- orders of magnitude
- control policies
- linear programming
- optimal solution
- similarity measure
- case study
- data mining
- dp matching
- real time