Michael P.Wellman. Market-oriented programming: Some early lessons. In Clearwater  Home/Search   Document Details and Download   Summary   Related Articles   Check

....this task was seen as an ideal candidate for a market based solution. In fact, since Reid Smith proposed his Contract Net Protocol [25] economically inspired mechanisms for coordination in multiagent systems have been widely studied and applied to a number of agent related problem domains [32]. Such mechanisms are attractive because they are inherently distributed, flexible, and in many cases quite scalable. Moreover, they are backed by a rich body of economic theory, so that rigorous results exists for many problems. Of the economicallybased control schemes, the one that has received ....

M. P. Wellman. Market-Oriented Programming: Some Early Lessons. In S. H. Clearwater, editor, Market-Based Control: A Paradigm for Distributed Resource Allocation, pages 74--95. World Scientific, Jan. 1996.

....Classification: 68M14,68T05 1998 ACM Computing Classification System: I.2.6,I.2.11 Keywords and Phrases: Reinforcement Learning, cooperative Multi Agent Systems Note: Work carried out under project SEN4 Evolutionary Systems and Applied Algorithmics . 1. Introduction As argued by Wellman [14, 15], a computational problem can be considered as a resource allocation problem. Borrowing from the insights of economics, it is however becoming increasingly clear that few concepts for resource allocation scale well with increasing complexity of the problem domain. In particular, centralized ....

M. P. Wellman. Market-oriented programming: Some early lessons. In S. Clearwater, editor, Market-Based Control: A Paradigm for Distributed Resource Allocation. World Scientific, River Edge, New Jersey, 1996.

....using classic Reinforcement Learning (RL) We further investigate several techniques from RL (model based learning, Q( # ) to scale application of the COIN framework. Lastly, the COIN framework is extended to improve performance for sequences of actions. 1 Introduction As argued by Wellman [14,15], a computational problem can be considered as a resource allocation problem. Borrowing from the insights of economics, it is however becoming increasingly clear that few concepts for resource allocation scale well with increasing complexity of the problem domain. In particular, centralized ....

M. P. Wellman. Market-oriented programming: Some early lessons. In S. Clearwater, editor, Market-Based Control: A Paradigm for Distributed Resource Allocation. World Scientific, River Edge, New Jersey, 1996.

....m i = s i , d i = A, v i = A, and p(S, S # ) 1 for all agents i and states S and S # . 3. TEAMINGANDMULTIAGENTSEARCH IN A TOD Common techniques to speed up multiagent search in taskoriented domains include communication, delegation (e.g. using contract net [8] and the use of auctions [9]. These and other grouping methods align an agent s desires with those of a larger team. However, we do not have any a priori evidence to make us believe that teaming will lead to better solutions for the system as a whole. Namely, we do not know whether the best solution, from a global ....

M. P. Wellman. Market-oriented programming: Some early lessons. In S. Clearwater, editor, Market-Based Control: A Paradigm for Distributed Resource Allocation. World Scientific, 1996.

....After more than a decade s development on e commerce, many kinds of logistics management models have been proposed and implemented. Some models are standalone and centralized, while others use a client server approach( 3, 4, 5] In recent years, researchers have proposed multi agent based models ([6, 7, 8, 9, 10]) However, most of the models regard logistics management as an auction, in which each entity tries to maximize its own bene t. Such an approach is appropriate for inter organizational logistics, which emphasizes maximizing individual pro t since suppliers and consumers belong to di erent ....

M.P. Wellman. Market-oriented programming: Some early lessons. In S. Clearwater, editor, Market-Based Control: A Paradigm for Distributed Resource Allocation, chapter 4. World Scienti c, 1996.

....a market clearing price, p , such that: # L l p d 0 (5) The new clearing price is fed back to the economic agents in an iterative way until the system has reached an equilibrium. The auction is then said to be complete. Further information about the auction process can be found in [17]. 5. Multi agent Architecture Given the computational economy defined in the previous section, the task of agentifying the system now comes into focus. This involves a mapping between the algorithmic view and the agent entity view of the system; in other words, viewing the Note that in ....

M. P. Wellman, Market-oriented Programming: Some Early Lessons, in S. Clearwater Ed., Market-based Control, World Scientific Publishing, 1996.

....the coordination mechanism can be distinguished according to the types mentioned above. Computable general equilibrium models [31] apply no dynamic mechanism at all, but calculate results out of total, global knowledge. The WALRASIAN auctioneer is directly implemented in marketoriented programming [38]. In this article, we show how the third possibility, the decentralized self organization, can be implemented and leads to coherent economic coordination with predictable results. In the remainder of this article, we will first clarify some definitions and introduce a framework for building ....

....the spread between input and output prices, and thus its utility, the agent follows a certain negotiation strategy. Comparable automated negotiation efforts in multiagent systems can be found in the research context of agent mediated electronic commerce [10; 32] and market oriented programming [38]. Human negotiation uses parameters such as demand level, concession, and concession rate: A bargainer s demand level can be thought of as the level of benefit to the self associated with the current offer or demand. A concession is a change of offer in the supposed direction of the other ....

Wellman, M.P.: Market-Oriented Programming: Some Early Lessons. Clearwater, S.H. (ed.). Market-Based Control: A Paradigm for Distributed Resource Allocation, 74-95. Singapore: World Scientific, 1996.

.... of Java [23] and distinguishable by a unique identity (ID) Each agent runs in parallel as single Java thread, and will not be synchronized explicitly as in common market implementations, which use time slicing or scheduling through arbitrators or auctioneers, as in Market Oriented Programming [32]. The AvInfoServer class has no effect on the market. At the moment, all relevant transaction data is reported from the trader agents to the AvInfoServer andthentoacentral log file, which can be visualized afterwards with common analyzing tools. 3.2. Individual Strategies The software agents ....

....an iterated contract net [30] and modified to reflect self interested behavior [6] 27] of the single software agent. Comparable automated negotiation efforts in multi agent systems can be found in the research context of agent mediated electronic commerce [2] 25] and market oriented programming [32]. The agent s set of actions consists of buying, selling, producing, moving, and termination. Depending on the market situation and its internal state, the software agent decides autonomously what to do. The basic algorithm works as follows: If the software agent has produced output in stock, it ....

Wellman, M.P., "Market-Oriented Programming: Some Early Lessons", in C learwater, S. (Ed.), Market-Based Contro l: A Paradigm for Distributed Resource Allocation,World Scientific, Singapore, 1996 .

....a certain degree of direct interaction between the agents is desirable but not necessary. The purpose of the interaction can be either competitive or co operative and the approach scales rather well and can thus be used for large agent systems as well. Market based Market Oriented Programming (WELLMAN, 1996) is the attempt to capture problems of distributed resource allocation with little information. The major advantage of this approach is that the agents need not provide their internal information to external authorities as the only information that is exchanged is the price. Market based ....

WELLMAN, M. P. (1996). Market-Oriented Programming: Some Early Lessons. In CLEARWATER, S. H., editor, Market-based Control. World Scientific.

....between activities [Malone Crowston 1990] given either a resource to be shared or a timing interdependency. The coordination mechanism can either be mediated (centralized, e.g. auctions) or unmediated with the use of subsequent bilateral or multilateral cooperation between all agents [Wellman 1996]. Examples for an unmediated bilateral mechanism in a value chain context are concepts for the coordination of manufacturing processes (e.g. KANBAN) where produced goods are sequentially and independently forwarded between autonomous workgroups as input factors for the next value chain step. ....

.... agents to get a full overview of any current market situation, this research project has many connections to the work of Sandholm concerning automated negotiation among self interested computationally limited agents [Sandholm 1996] We also benefit from the market oriented programming approach of Wellman [Wellman 1996], especially in the design of our goods and mechanism space. But in contrast to the works of Sandholm, Wellman and Jennings [Jennings 1996] our agents are neither intelligent nor do they act strategically. Their actions are based on stochastic operations on the initial and never changing ....

Wellman, Michael P.: Market-Oriented Programming: Some Early Lessons, in: Clearwater (Ed.): Market-Based Control: A Paradigm for Distributed Resource Allocation, World Scientific. 1996.

....complementarities is described in [20] Wellman et al. discuss the problems of trying to apply Walrasian like economic coordination models of competitive equilibrium in a non convex domain. Wellman also gives an interesting motivational account of his experience with Market oriented Programming in [19]. A recent report of Tuomas Sandholm [16] suggests an efficient algorithm for winner determination in combinatorial auctions and gives pointer to further relevant literature. Especially the problem of NP hardness and weak approximability is discussed. A similar approach is presented in [6] A ....

M. Wellman. Market-oriented programming: Some early lessons. In S. Clearwater, editor, Market-Based Control. World Scientific, Singapore, 1996.

....resource allocation via computational markets yields inherent survivability hinges on the premise that most important applications can be cast in this framework. We have significant experience in mapping a wide range of problems to the market framework, including those described here and others (Wellman 1996). Most practical resource allocation problems fall in one of the four categories formed by crossing discrete or continuous quantity scales with discrete or continuous time scales. Therefore, many interesting problems fit models we have already designed and tested. In this section we describe two ....

Wellman, Michael P. 1996. Market-oriented programming: Some early lessons. In Market-Based Control: A Paradigm for Distributed Resource Allocation, edited by S. Clearwater: World Scientific.

....2.2 suggest that under some circumstances centralized design may yield few benefits, because decentralized computation may lead to equally good outcomes. More sophisticated results of a similar flavor can be found in the literature on microeconomic approaches to distributed resource allocation [38]. The main contribution of Section 3 is to present a fast simultaneous simulation technique adopted from the processormemory literature but more suitable to Web related research, and demonstrate its application to the optimal cache sizing problem. The idealized model of Section 2 is useful for ....

M.P. Wellman, Market-oriented programming: some early lessons, in: S. Clearwater (Ed.), Market-Based Control: A Paradigm for Distributed Resource Allocation, World Scientific, Singapore, 1996, http://ai.eecs.umich.edu/people/wellman/Publications.html.

....any period. We have also successfully coupled this approach with learning, and although learning in multiagent systems is often hard, here it is strategyfree, complete, and convergent. Comparison With Other Work Wellman has used market oriented programming to solve various distributed problems (Wellman 1996). He assumes that agents act as price takers even in small MAS when a strategic agent would be expected to outperform such a naive strategy (Vidal Durfee 1996; Hu Wellman 1996) This price taking assumption is only rational for self interested agents when the size of each agent is small in ....

.... Comparison With Other Work Wellman has used market oriented programming to solve various distributed problems (Wellman 1996) He assumes that agents act as price takers even in small MAS when a strategic agent would be expected to outperform such a naive strategy (Vidal Durfee 1996; Hu Wellman 1996). This price taking assumption is only rational for self interested agents when the size of each agent is small in comparison with the total system. The joint plans that result are also only efficient in the absence of externalities. The Compensation Mechanism allows for strategic agents and ....

Wellman, M. P. 1996. Market-oriented programming: Some early lessons. In Clearwater, S. H., ed., MarketBased Control. World Scientific. chapter 4, 74--95.

No context found.

Michael P.Wellman. Market-oriented programming: Some early lessons. In Clearwater

....methodology for computational markets, however, requires an analytical characterization of their properties. In our own MOP work, we have adopted the framework of general equilibrium theory [20] and have found that our computational markets behave predictably when conditions of the theory are met [24, 39, 41]. We have also applied the approach to discrete optimization problems where the conditions guaranteeing desirable outcomes are not satisfied and have found (not surprisingly) that the methods sometimes work, and other times break down [38, 40] Since scheduling problems very often involve ....

Michael P. Wellman. Market-oriented programming: Some early lessons. In Clearwater [6].

No context found.

Wellman, M.P.: Market-oriented programming: Some early lessons. In Clearwater, S., ed.: Market-Based Control: A Paradigm for Distributed Resource Allocation. World Scientific (1996)

No context found.

M. P. Wellman. Market-oriented programming: some early lessons. In S. Clearwater, editor, Market-Based Control: A Paradigm for Distributed Resource Allocation, HonkKong, equilibrium problem. European Journal of Operational Research, 83:117-136, 1995.

No context found.

Wellman MP. Market-oriented programming: some early lessons. In: Clearwater S,editor. Market-based control: A paradigm for distributed resource allocation. Hong-Kong: World Scientific Publishers; 1996.

No context found.

M. P. Wellman. Market-oriented programming: some early lessons. In S. Clearwater, editor, MarketBased Control: A Paradigm for Distributed Resource Allocation, Honk-Kong, 1996. World Scientic. 22

No context found.

M. P. Wellman. Market-oriented programming: some early lessons. In S. Clearwater, editor, MarketBased Control: A Paradigm for Distributed Resource Allocation, Honk-Kong, 1996. World Scientic. 22

No context found.

Wellman, M., "Market Oriented Programming: Some Early Lessons", In S.H.Clearwater (ed), Market-Based Control a Paradigm for Distributed Resource Allocation, World Scientific Press, 1996.

No context found.

M. P. Wellman. Market-oriented programming: Some early lessons. In S. Clearwater, editor, Market-Based Control: A Paradigm for Distributed Resource Allocation. World Scientific, 1996.

No context found.

Michael P Wellman. Market-oriented programming: Some early lessons. In Clearwater [Cle96], chapter 4, pages 74--95.

No context found.

: Market-Based Control: A Paradigm for Distributed Resource Allocation. World Scientific, 1996. ftp://ftp.eecs.umich.edu/people/wellman/mbc95.p s.Z

First 50 documents  Next 50

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