Analyzing and visualizing multiagent rewards in dynamic and stochastic domains

By Adrian K. Agogino, Kagan Tumer. 2008

129 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

Benchmarking Multi-Agent Deep Reinforcement Learning Algorithms in Cooperative Tasks (2020, background): “it simplifies the learning objective for each agent to maximise, while ensuring the global objective is maximised by all agents [Agogino and Tumer, 2008].”
Multiagent Learning: Basics, Challenges, and Prospects (2012, background): “…learn to behave optimally in the presence of other (learning) agents, facing the complexities of incomplete information, large state spaces, credit assignment, cooperative or competitive settings, and reward shaping (Wolpert and Tumer 2001; Tumer, Agogino, and Wolpert 2002; Agogino and Tumer 2008).”
Multiagent Bidirectionally-Coordinated Nets for Learning to Play StarCraft Combat Games (2017, background): “Reward shaping is of great importance in learning the collaborative or competing behaviours among agents [40].”
Comparative Evaluation of Multi-Agent Deep Reinforcement Learning Algorithms (2020, background): “it simplifies the learning objective for each agent to maximise, while maintaining the factorisation to ensure the global objective is maximised by all agents [1].”
Local Approximation of Difference Evaluation Functions (2016, background): “First, they are aligned with the overall system objective [2].”
When Does Communication Learning Need Hierarchical Multi-Agent Deep Reinforcement Learning (2019, background): “Since coordination is the task ’ s main objective, we chose a completely shared reward (Agogino and Tumer 2008).”
Combining reward shaping and hierarchies for scaling to large multiagent systems (2016, unknown): “Reward shaping has been shown to drastically improve coordination, scalability, and performance in multiagent systems (Agogino & Tumer, 2008; Williamson et al ., 2009; Grzes & Kudenko, 2010).”
Modeling difference rewards for multiagent learning (2012, background): “Difference rewards have been shown to perform very well in multiagent domains [1].”