Learning Incentive Design Policies in Orienteering Games
Dnr:
NAISS 2025/22-677
Type:
NAISS Small Compute
Principal Investigator:
Malintha Fernando Chakravarthige
Affiliation:
Kungliga Tekniska högskolan
Start Date:
2025-05-02
End Date:
2026-06-01
Primary Classification:
10201: Computer Sciences
Allocation
- Klemming at PDC: 500 GiB
- Mimer at C3SE: 500 GiB
- Alvis at C3SE: 250 GPU-h/month
- Dardel-GPU at PDC: 200 GPU-h/month
- Dardel at PDC: 1 x 1000 core-h/month
Abstract
This project is based on the digital futures postdoc project titled "Designing interaction-aware multi-robot systems". In this project, we aim to learn decentralized reinforcement learning policies for heterogeneous robotic teams comprising individualistic agents to complete tasks in complex, dynamic environments.