Layered learning in multiagent systems - a winning approach to robotic soccer

By Peter Stone. Intelligent robotics and autonomous agents, 2000

345 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/2b57e0525a4babf758bf4a959aac5086f1097a71

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

Evolving Soccer Keepaway Players Through Task Decomposition (2005, methodology): “Layered learning was originally developed for the complex, multiagent learning task of simulated robot soccer (Noda et al., 1998; Stone, 2000).”
Unifying temporal and structural credit assignment problems (2004, background): “Robotic soccer is a well studied domain that clearly exhibits this form of credit assignment problem, where learning algorithms need to judge a particular player’s role in achieving the overall goal of the mutli-agent system of winning the game [6].”
From Motor Control to Team Play in Simulated Humanoid Football (2021, methodology): “In Layered Learning a bottom-up hierarchical task decomposition is given and implemented architecturally, with the output of one layer feeding into the next.”
Multiagent Systems: A Survey from a Machine Learning Perspective (2000, methodology): “Layered learning [Stone, 2000] has been implemented in the simulated robotic soccer domain.”
Making friends on the fly: Cooperating with new teammates (2017, unknown): “Email addresses: sam@cogitai.com (Samuel Barrett), rosenfa@jct.ac.il (Avi Rosenfeld), sarit@cs.biu.ac.il (Sarit Kraus), pstone@cs.utexas.edu (Peter Stone) 1This work was performed while Samuel Barrett was a graduate student at the University of Texas at Austin.”
Scaling Reinforcement Learning toward RoboCup Soccer (2001, background): “PassBall(k), however, requires an extended sequence of actions—getting the ball into position for kicking, and then a sequence of kicks in the desired direction (Stone, 2000); a single invocation of PassBall(k) influences behavior for several time steps.”
Handbook of Granular Computing (2008, methodology): “Two problems have been particularly addressed: (i) for the solution of fuzzy algebraic equations, in particular having the matrix form Au = v with some or all elements being fuzzy (A, v, and/or u) (see [74–78]); (ii) the analysis and numerical treatment of fuzzy integral (see [79]) and differential ”
Evolving Keepaway Soccer Players through Task Decomposition (2003, methodology): “Such task decomposition is a powerful, general principle in artificial intelligence that has been used successfully with machine learning in the full robotic soccer task [12].”