• Corpus ID: 246867119

Planning Spatial Networks with Monte Carlo Tree Search

@inproceedings{Darvariu2021PlanningSN,
  title={Planning Spatial Networks with Monte Carlo Tree Search},
  author={Victor-Alexandru Darvariu and Stephen Hailes and Mirco Musolesi},
  year={2021}
}
We tackle the problem of goal-directed graph construction: given a starting graph, a budget of modifications, and a global objective function, the aim is to find a set of edges whose addition to the graph achieves the maximum improvement in the objective (e.g., communication efficiency). This problem emerges in many networks of great importance for society such as transportation and critical infrastructure networks. We identify two significant shortcomings with present methods. Firstly, they… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 57 REFERENCES
Goal-directed graph construction using reinforcement learning
TLDR
This work formulate the construction of a graph as a decision-making process in which a central agent creates topologies by trial and error and receives rewards proportional to the value of the target objective.
Shape and efficiency in spatial distribution networks
TLDR
This work proposes two models of network growth that offer explanations of how this situation might arise and finds that distribution networks appear remarkably close to optimal where both these properties are concerned.
Upgrading Shortest Paths in Networks
TLDR
The Upgrading Shortest Paths Problem is introduced, a new combinatorial problem for improving network connectivity with a wide range of applications from multicast communication to wildlife habitat conservation and it is shown that this problem is NP-hard.
Algebraic connectivity optimization via link addition
TLDR
This work proposes two strategies to optimize a network for a given dynamic process via a minor topological modification, and suggests that the Fiedler vector based strategy 2 performs better than strategy 1, i.e. node degree.
Optimizing Resilience in Large Scale Networks
TLDR
This work proposes a decision making framework to optimize the resilience of road networks to natural disasters such as floods and presents a fast algorithm based on the sample average approximation (SAA) method and network design techniques to solve this problem approximately.
Spatial Networks
  • M. Barthelemy
  • Computer Science
    Encyclopedia of Social Network Analysis and Mining
  • 2014
Neural Combinatorial Optimization with Reinforcement Learning
TLDR
A framework to tackle combinatorial optimization problems using neural networks and reinforcement learning, and Neural Combinatorial Optimization achieves close to optimal results on 2D Euclidean graphs with up to 100 nodes.
Pre-disaster investment decisions for strengthening a highway network
A long-time limit for world subway networks
TLDR
It is shown that the temporal evolution of the structure of the world's largest subway networks converge to a shape that shares similar generic features despite their geographical and economic differences, and suggests the existence of dominant, universal mechanisms governing the evolution of these structures.
...
...