# Dynamic neighbourhood optimisation for task allocation using multi-agent

arXiv: https://arxiv.org/abs/2102.08307 · HTML: https://mlsanity.com/paper/2102.08307/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 103 references that are in the map.
- First posted on arXiv 16 Feb 2021.

## Builds on (most cited first, 11 of 11)

- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Selective Experience Replay for Lifelong Learning](https://mlsanity.com/paper/1802.10269/) (2018)
- [Toward Understanding Catastrophic Forgetting in Continual Learning](https://mlsanity.com/paper/1908.01091/) (2019)
- [Near Optimal Behavior via Approximate State Abstraction](https://mlsanity.com/paper/1701.04113/) (2017)
- [Optimal Exploration-Exploitation in a Multi-Armed-Bandit Problem with Non-stationary Rewards](https://mlsanity.com/paper/1405.3316/) (2014)
- [No-Regret Exploration in Goal-Oriented Reinforcement Learning](https://mlsanity.com/paper/1912.03517/) (2019)
- [Bayesian Curiosity for Efficient Exploration in Reinforcement Learning](https://mlsanity.com/paper/1911.08701/) (2019)
- [Reward Design in Cooperative Multi-agent Reinforcement Learning for Packet Routing](https://mlsanity.com/paper/2003.03433/) (2020)
- [Overcoming Catastrophic Interference in Online Reinforcement Learning with Dynamic Self-Organizing Maps](https://mlsanity.com/paper/1910.13213/) (2019)
- [Organization of Multi-Agent Systems: An Overview](https://mlsanity.com/paper/1506.09032/) (2015)
- [Multi-Agent Deep Reinforcement Learning enabled Computation Resource Allocation in a Vehicular Cloud Network](https://mlsanity.com/paper/2008.06464/) (2020)

## Built on by (most cited first, 1 of 1)

- [A Survey on Large-Population Systems and Scalable Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2209.03859/) (2022)

## Abstract (arXiv)

In large-scale systems there are fundamental challenges when centralised techniques are used for task allocation. The number of interactions is limited by resource constraints such as on computation, storage, and network communication. We can increase scalability by implementing the system as a distributed task-allocation system, sharing tasks across many agents. However, this also increases the resource cost of communications and synchronisation, and is difficult to scale. In this paper we present four algorithms to solve these problems. The combination of these algorithms enable each agent to improve their task allocation strategy through reinforcement learning, while changing how much they explore the system in response to how optimal they believe their current strategy is, given their past experience. We focus on distributed agent systems where the agents' behaviours are constrained by resource usage limits, limiting agents to local rather than system-wide knowledge. We evaluate these algorithms in a simulated environment where agents are given a task composed of multiple subtasks that must be allocated to other agents with differing capabilities, to then carry out those tasks. We also simulate real-life system effects such as networking instability. Our solution is shown to solve the task allocation problem to 6.7% of the theoretical optimal within the system configurations considered. It provides 5x better performance recovery over no-knowledge retention approaches when system connectivity is impacted, and is tested against systems up to 100 agents with less than a 9% impact on the algorithms' performance.