Stone, P. and Veloso, M. M.: 1998, Towards collaborative and adversarial learning: A case study in robotic soccer, Internat. J. Human--Comput. Systems 48.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....references we also summarize the significant results of these teams in international competitions (top 10 finishes only) It is important to realize that the list is not exhaustive and only contains references and results up to and including the year 2000. Team References Roll of honour CMUnited [7, 90, 92, 93, 94, 95, 96, 97] 4th WC97, 1st WC98, 98, 99, 100, 102, 103, 104, 115] 1st 9th WC99, 3rd 4th WC00 Essex Wizards [36, 37, 39, 50, 51, 52, 53] 3rd WC99, 3rd EC00, 7th WC00 FC Portugal [56, 76, 77] 1st EC00, 1st WC00 Cyberoos [13, 69, 70, 71, 72, 73, 74] 3rd PR98, 4th EC00, 9th WC00 Karlsruhe Brainstormers [79, ....

P. Stone and M. Veloso. Towards Collaborative and Adversarial Learning: A Case Study in Robotic Soccer. International Journal of Human Computer Studies, 48(1):83--104, Jan. 1998.

....outperforms the basic planner in a robot domain. Keywords: Planning, unsupervised machine learning, autonomous intelligent systems, theory formation and revision. 1 Introduction Autonomous intelligent behavior is an area with an emerging interest within Artificial Intelligence researchers [6, 10, 13, 14]. It integrates many areas, such as robotics, planning, and machine learning. This integration opens many questions that arise when designing such systems, such as how operator descriptions can be incrementally and automatically acquired from the planning execution cycle, or how a planner can use ....

Peter Stone and Manuela M. Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. To appear in International Journal of HumanComputer Systems (IJHCS), 1996.

....layer Agents select their formation, tactics, and decide the policy of resource management. These must be decided depending upon opponent model. 2] Static role assignment is done in this layer. There are many team styles, indeed. So we need to adapt them effectively. Layer approach is popular one[4][5] 4.2 Group Layer Group Layer planning include about three or four teammates in local state near ball. In group layer agent who finds the chance, can be a planner. If there are no fatal condition to execute the plan, agreement will be done, and plan in group level can be executed. 156 4.3 ....

Peter Stone, Manuela Veloso. Towards Collaborative and Adversarial Learning A Case Study in Robotic Soccer. . IJHCS. 1998.

.... has been successfully applied in communication free settings (Mataric, 1992; Sen, Sekaran, and Hale 1994; Sen and Sekaran 1998) and has even been used to learn a very simple communication protocol (Yanco Stein, 1993) Another approach to learning to control reactive behaviors can be found in Stone Veloso (1998). 30 6 Experimental Analysis As the members of the community of actors become familiar with the capabilities of other members of the community and the regular problems of coordination that exist in their domain of activity, behavioral conventions begin to emerge. As actors proceed through their ....

Stone, P. and Veloso, M. (1998). Towards collaborative and adversarial learning: a case study in robotic soccer. International Journal of Human-Computer Studies, 48:83--104.

.... has been successfully applied in communication free settings (Mataric, 1992; Sen, Sekaran, and Hale 1994; Sen and Sekaran 1998) and has even been used to learn a very simple communication protocol (Yanco Stein, 1993) Another approach to learning to control reactive behaviors can be found in Stone Veloso (1998). 23 3.4 Communication MOVERS WORLD, and the techniques used to learn coordinated procedures by MOVERSWORLD actors, is shaped by the fact that communication is the central mechanism for cooperation and coordination. Communication is considered expensive , so it occurs at runtime and is never ....

Stone, P. and Veloso, M. (1998). Towards collaborative and adversarial learning: a case study in robotic soccer. International Journal of Human-Computer Studies, 48:83--104.

....surrounding soccer agent and receives control commands to have an agent act several times a second. Learning for soccer agents and soccer robots is one of the most important issues in RoboCup, and some studies have been reported, such as learning of cooperative behavior with neural network [8], acquiring shoot techniques for real E mail: tohgoroh ics.nitech.ac.jp Fig. 1. An image of the virtual eld which Soccer Server provides. robots with reinforcement learning [9] and learning of position arrangements with genetic algorithms [10] In this paper, we propose a framework for ....

Stone, P. and Veloso, M.: \Towards Collaborative and Adversarial Learning: A Case Study in Robotic Soccer," Proc. International Journal of Human-Computer Systems, 48, (1998).

....with partial and uncertain information received from the RoboCup server. 2.5. 2 Machine Learning In another software soccer server which closely models the real world robotic soccer competition, Stone and Veloso use a neural network to teach a robot how to kick a moving ball into an open goal [SV98b] After training the network to shoot the ball with the robot beginning in the upperleft quadrant of the field, the robot was moved to the three other quadrants and tested with the learned behaviour. The behaviour generalised well to the different situations, shooting the ball into the goal 96.5 ....

....behaviours provided in experiment 3. As mentioned in the analysis of that experiment (section 3.4.6) a soccer player using a high level function set is only as good as the functions provided. While Luke et al. state that behaviours such as ball interception are difficult to evolve, in both [SV98b] and [SV98a] Stone and Veloso have shown that such behaviours can be learned by a neural network, and that they are often just as robust, if not more so, than hand coded approaches. It would be interesting to mirror the machine learning work conducted by Stone and Veloso with an approach which ....

Peter Stone and Manuela Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. International Journal of Human-Computer Studies, 48(1):83--104, 1998.

....player, the field of machine learning and the study of game playing have come together to yield several significant advances. To date, research in multiple agent planning and control has been limited largely to the area of distributed artificial intelligence (Rosenschein and Genesereth 1985; Stone and Veloso 1996a; Suguwara and Lesser 1993; Tan 1993) and artificial life (Collins 1992; Huberman and Glance 1995; Sandholm and Crites 1995; Stanley, Ashlock, and Tesfastsion 1993) In distributed AI (DAI) several agents cooperate to achieve some goal or accomplish some task. The task is usually one of ....

.... on individual behaviors and artificial life focuses on population dynamics (Collins 1992) So far, most work in learning and multi agent systems has emphasized multiple agents learning complementary behaviors in a coordinated environment to accomplish some task, such as 2 team game playing (Stone and Veloso 1996b; Tambe 1996a) combinatorial optimization (Dorigo, Maniezzo, and Colorni 1996) and obstacle avoidance (Grefenstette 1991) The research discussed in this paper focuses on exploring methods of learning in the context of competitive multi agent systems. In particular, we focus on exploring ....

[Article contains additional citation context not shown here]

Stone, P. and Veloso, M. (1996b). Towards collaborative and adversarial learning: A case study in robotic soccer. In 1996 AAAI Spring Symposium on Adaptation, Co-Evolution, and Learning in Multiagent Systems.

....constructing internal state spaces representing the environment. Other soccer playing robots are being developed by Sahota [14, 13, 12] and Shimada et al. 15] Also, an earlier version of the Soccer Server has been used by Stone and Veloso to investigate learning under multi agent environments [16, 17]. They apply techniques of neural networks and machine learning to improve a player s skills and to acquire the ability to select good play plans. 2.2 RoboCup The World Cup Robot Soccer Initiative (RoboCup) is an attempt to foster AI and intelligent robotics research by providing a standard ....

Peter Stone and Manuela Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. Proceedings SS-96-01, pp. 88--92, 1996 AAAI Symposium, March 1996.

.... scalability of the underlying model checking representation and search techniques, it can be shown to be a very efficient non deterministic planner [9, 10] One of our main research objectives is to develop planning systems suitable for planning in uncertain, single, or multi agent environments [25, 42, 39]. The universal planning approach, as originally developed [38] is appealing for this type of environments. A universal plan is a set of state action rules that aim at covering the possible multiple situations in the non deterministic environment. A universal plan is executed by interleaving the ....

P. Stone and M. M. Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. International Journal of Human-Computer Studies (IJHCS), 1998.

.... 4 Implementation in Robotic Soccer Robotic soccer is a very good example of a PTS domain: teams can coordinate before the game, at half time, and at other break points, but communication is limited during play [10, 12] Robotic soccer systems have been recently developed both in simulation [14, 24, 25] and with real robots [1, 9, 19, 20, 28] The research presented in this paper was first developed in simulation and it has also been successfully used on our real robot team. The soccer server [16] version 3 of which serves as the substrate simulator for the research reported in this paper, ....

Peter Stone and Manuela Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. International Journal of Human-Computer Studies, 48(1):83--104, January 1998.

.... Robotic Soccer Robotic soccer is a very good domain for studying real time multi agent coordination techniques: agents must act quickly and autonomously while contributing to the achievement of the team s overall goal [4, 7] Robotic soccer systems have been recently developed both in simulation [9, 14, 15] and with real robots [1, 3, 12, 13, 20] The research presented in this paper was developed jointly in simulation and on our real robot team. 2.1 Simulator The RoboCup soccer server [10] has been used as the basis for successful international competitions [11] and research challenges [5] Though ....

Peter Stone and Manuela Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. International Journal of Human-Computer Studies, 48(1):83--104, January 1998.

....the ball, to arbitrarily complex reasoning procedures that take into account the actions and perceived strategies of teammates and opponents. Opportunities, and indeed demands, for innovative and novel techniques abound. Robotic Soccer systems have been recently developed both in simulation [6, 9, 12, 14] and with real robots [1, 4, 10, 11, 15, 13] While robotic systems are difficult, expensive, and time consuming to use, they provide a certain degree of realism that is never possible in simulation. On the other hand, simulators allow researchers to isolate key issues, implement complex ....

Peter Stone and Manuela Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. in International Journal of Human-Computer Systems (IJHCS), 48, 1998.

....even know their neighboring states. Thus unlike TPOT RL agents, the nodes are able to use dynamic programming. In other soccer systems, there have been a number of learning techniques that have been explored. However, most have learned low level, individual skills as opposed to team based policies [1, 10]. Interestingly, 6] uses genetic programming to evolve team behaviors from scratch as opposed to our layered learning approach. TPOT RL is an adaptation of RL to non Markovian multi agent domains with opaque transitions, large state spaces, hidden state and limited training opportunities. The ....

Peter Stone and Manuela Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. International Journal of Human-Computer Studies, 48(1):83--104, January 1998.

....robotic soccer is a great domain for multiagent Machine Learning. In another soccer simulator, Stone and Veloso use Memory based Learning to allow a player to learn when to shoot and when to pass the ball [78] They then use Neural Networks to teach a player to shoot a moving ball into the goal [79]. They use similar techniques in the soccerserver system as well, extending the learned behavior as a part of a hierarchical learning system [80] Matsubar et al. also use a Neural Network to allow a player to learn when to shoot and when to pass in the soccerserver system [81] Once low level ....

P. Stone and M. M. Veloso, "Towards collaborative and adversarial learning: A case study in robotic soccer," To appear in International Journal of Human-Computer Systems (IJHCS), 1997.

No context found.

Peter Stone and Manuela M. Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. To appear in International Journal of HumanComputer Systems (IJHCS), 1997.

....in simulation and with real robots. Using a simulator based closely upon the Dynasim system [28] we previously used Memory based Learning to allow a player to learn when to shoot and when to pass the ball [31] We then used Neural Networks to teach a player to shoot a moving ball into the goal [35]. In the soccer server, we then layered two learned behaviors to produce a higher level multi agent behavior: passing [34] Also in the soccer server Matsubara et al. used a Neural Network to allow a player to learn when to shoot and when to pass [23] as opposed to the Memory based technique ....

Peter Stone and Manuela Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. International Journal of Human-Computer Studies, 48(1):83--104, January 1998.

....knowledge. Between exist several variations with more or less knowledge. The approaches are summarized as follows: 1) complete hand coding (no learning) 2) parameter tuning given the structural (qualitative) knowledge (self calibration) 3) Subtask learning fit together in a layered fashion [ Stone and Veloso, 1997 ] 4) typical reinforcement learning such as Q learning with almost no a priori knowledge, but given the state and action spaces, 5) action selection from the state and action space construction [ Asada et al. 1996a, Takahashi et al. 1996 ] and (6) tabula rasa learning. These approaches ....

....In addition to issues in the challenge (1) several other issues remain: ffl Situation A: Prediction of the ball speed and direction is a key issue to receive the ball. To receive the passed ball while moving, the relationship between the moving ball and the self motion should be made clear [ Stone and Veloso, 1997 ] ffl Situation B: In addition to the above issue, goal protection is important. To estimate the goal position, the agent may have to watch the goal area lines and the penalty area line. Again, the omni directional lens is much better to see the coming ball and the goal position simultaneously. ....

P. Stone, M. Veloso. Towards Collaborative and adversarial learning: A case study in robotic soccer . In International Journal of Human-Computer Systems (IJHCS), 1997.

No context found.

Stone, P. and Veloso, M. M.: 1998, Towards collaborative and adversarial learning: A case study in robotic soccer, Internat. J. Human--Comput. Systems 48.

No context found.

Peter Stone and Manuela Veloso. Towards collaborative and adversarial learning: A case study in robotic soccer. Technical report, School of Computer Science. CMUCS -95-207. Carnegie Mellon University, 1995.

No context found.

Peter Stone and Manuela Veloso., `Towards collaborative and adversarial learning: A case study in robotic soccer', International Journal of Human-Computer Systems, 48(1), (1998).

No context found.

Peter Stone and Manuela Veloso. Toward collaborative and adversarial learning: A case study in robotic soccer. International Journal of Human Computer Studies, 48, 1998.

No context found.

P. Stone and M. Veloso. Towards collaborative and adversarial learning: a case study in robotic soccer. In S. Sen, editor, AAAI Spring Symposium on Adaptation, Coevolution and Learning in multi-agent systems, March 1996.

No context found.

P. Stone and M. Veloso. Towards collaborative and adversarial learning: a case study in robotic soccer. In S. Sen, editor, AAAI Spring Symposium on Adaptation, Coevolution and Learning in multi-agent systems, March 1996.

No context found.

Stone, P., and Veloso, M. 1996b. Towards Collaborative and Adversarial Learning: A Case Study in Robotic Soccer. International Journal of Human-Computer Studies/Knowledge Acquisition.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC