Using communication to reduce locality in distributed multiagent learning

By Maja J. Mataric. 1997

How researchers see this paper

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

Citations in the corpus

Stabilising Experience Replay for Deep Multi-Agent Reinforcement Learning (2017, background): “However, they also argue that the simple estimation of the value function in the standard model-free fashion is not enough to solve multi-agent problems, and coordination through means such as communication (Mataric, 1998) is required to ground separate observations to the full state function.”
Independent reinforcement learners in cooperative Markov games: a survey regarding coordination problems (2012, background): “Various RL algorithms have been proposed in the literature for ILs and fully-cooperative MAS.”
Half Field Offense in RoboCup Soccer: A Multiagent Reinforcement Learning Case Study (2006, background): “Mataric [7] uses reinforcement learning to train real robots on a box-pushing task.”
Imitation and Reinforcement Learning in Agents with Heterogeneous Actions (2001, background): “The ability of imitation to effect skill transfer between agents has bee n demonstrated in a range of domains (Atkeson & Schaal, 1997; Billard & Hayes, 1997; Hayes & Demiris, 1994; Kuniyoshi et al., 1994; Mataric, 1998; Mitchell et al., 1985; Utgoff & Clouse, 1991).”
Awareness coverage control over large scale domains with intermittent communications (2008, background): “In [24], local undirected communication is used in fully distributed multi-agent systems.”
Awareness Coverage Control Over Large-Scale Domains With Intermittent Communications (2010, background): “Both [30] and [31] demonstrate improvements in global behavior made by exchanging local sensing information.”
Swarm robotics: a review from the swarm engineering perspective (2013, methodology): “Matari´c and her colleagues addressed this problem using communication (Matari´c 1998) or behavioral decomposition (Matari´c 1997).”
An Overview of Cooperative and Competitive Multiagent Learning (2005, background): “For example, agents may use higherlevel descriptions of states and actions [104].”
Accelerating Reinforcement Learning through Implicit Imitation (2003, methodology): “n expert mentor. This type of learning can be brought about through explicit teaching or demonstration (Atkeson & Schaal, 1997; Lin, 1992; Whitehead, 1991a), by sharing of privileged information (Mataric, 1998), or through an explicit cognitive representation of imitation (Bakker & Kuniyoshi”
Learning in multiagent systems (1999, methodology): “Related work that focuses on multirobot learning was presented, e.g., by Matari c [31, 32 ] and Parker [38, 39].”
Cooperative Multi-Agent Learning: The State of the Art (2005, background): “For example, Mataric [143, 140] applies Q learning to select from hand-coded reactive behaviors such as avoid, head-home, search or disperse for robot foraging tasks.”