A comparison of Monte Carlo tree search and rolling horizon optimization for large-scale dynamic resource allocation problems

2017 ◽  
Vol 263 (2) ◽  
pp. 664-678 ◽  
Author(s):  
Dimitris Bertsimas ◽  
J. Daniel Griffith ◽  
Vishal Gupta ◽  
Mykel J. Kochenderfer ◽  
Velibor V. Mišić
Keyword(s):  
Resource Allocation ◽  
Monte Carlo ◽  
Large Scale ◽  
Dynamic Resource Allocation ◽  
Tree Search ◽  
Rolling Horizon ◽  
Monte Carlo Tree Search ◽  
Allocation Problems ◽  
Dynamic Resource ◽  
Rolling Horizon Optimization

A population game approach for dynamic resource allocation problems

2016 ◽  
Vol 90 (9) ◽  
pp. 1957-1972 ◽  
Author(s):  
Ashkan Pashaie ◽  
Lacra Pavel ◽  
Christopher J. Damaren
Keyword(s):  
Resource Allocation ◽  
Dynamic Resource Allocation ◽  
Allocation Problems ◽  
Population Game ◽  
Dynamic Resource

Evolutionary Methods for Solving Dynamic Resource Allocation Problems

2019 ◽  
Vol 80 (7) ◽  
pp. 1335-1346
Author(s):  
G. I. Beliavsky ◽  
N. V. Danilova ◽  
G. A. Ougolnitsky
Keyword(s):  
Resource Allocation ◽  
Dynamic Resource Allocation ◽  
Evolutionary Methods ◽  
Allocation Problems ◽  
Dynamic Resource

scholarly journals Tackling Sparse Rewards in Real-Time Games with Statistical Forward Planning Methods

2019 ◽  
Vol 33 ◽  
pp. 1691-1698 ◽  
Author(s):  
Raluca D. Gaina ◽  
Simon M. Lucas ◽  
Diego Pérez-Liébana
Keyword(s):  
Monte Carlo ◽  
Evolutionary Algorithms ◽  
Real Time ◽  
Fitness Landscape ◽  
Tree Search ◽  
Rolling Horizon ◽  
Reward Systems ◽  
Monte Carlo Tree Search ◽  
Planning Methods ◽  
High Level

One of the issues general AI game players are required to deal with is the different reward systems in the variety of games they are expected to be able to play at a high level. Some games may present plentiful rewards which the agents can use to guide their search for the best solution, whereas others feature sparse reward landscapes that provide little information to the agents. The work presented in this paper focuses on the latter case, which most agents struggle with. Thus, modifications are proposed for two algorithms, Monte Carlo Tree Search and Rolling Horizon Evolutionary Algorithms, aiming at improving performance in this type of games while maintaining overall win rate across those where rewards are plentiful. Results show that longer rollouts and individual lengths, either fixed or responsive to changes in fitness landscape features, lead to a boost of performance in the games during testing without being detrimental to non-sparse reward scenarios.

Sign in / Sign up

Export Citation Format

Share Document