Tesauro, G., & Kephart, J. (1999). Pricing in agent economies using multi-agent Qlearning. Proceedings of Fifth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty (pp. 71--86).  Home/Search   Document Details and Download   Summary   Related Articles   Check

....from machine learning to explore the space of possible ways in which agents might act in particular markets. For example, reinforcement learning has been used to explore bidding patterns in auctions [20, 23] and establish the ways in which price setting behaviour can a ect consumer markets [27]. Another approach is to use techniques from evolutionary computing, that is from genetic algorithms [12] and genetic programming [16] Inspired by the biological metaphor of evolution, genetic algorithms code aspects of a solution to a problem in an arti cial chromosome (typically a binary ....

G. Tesauro and J. O. Kephart. Pricing in agent economies using multi-agent Qlearning. Autonomous Agents and Multi-Agent Systems, 5(3):289-304, 2002.

....from machine learning to explore the space of possible ways in which agents might act in particular markets. For example, reinforcement learning has been used to explore bidding patterns in auctions [20, 23] and establish the ways in which price setting behaviour can affect consumer markets [27]. Another approach is to use techniques from evolutionary computing, that is from genetic algorithms [12] and genetic programming [16] Inspired by the biological metaphor of evolution, genetic algorithms code aspects of a solution to a problem in an artificial chromosome (typically a binary ....

G. Tesauro and J. O. Kephart. Pricing in agent economies using multi-agent Qlearning. Autonomous Agents and Multi-Agent Systems, 5(3):289-304, 2002.

....seller uses available information about the market, such as the distribution of buyer preferences, or its competitor s prices. There has been recent work in the literature which attempt to address the question of automated dynamic pricing assuming more or less complete information about the market [6, 8, 12]. But what if the seller has only limited information about its environment In our earlier work, we have explored how a monopolistic seller might dynamically set its price schedule to maximize profit in a market where it has to learn the buyer preferences [2, 7] In this work, we study markets ....

G. J. Tesauro and J. O. Kephart. Pricing in agent economies using multi-agent q-learning. In Proceedings of Fifth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, 1999.

....discussion of the use and misuse of genetic algorithms, genetic programming, and other forms of evolutionary learning representations in the modeling of social processes. Additional types of learning algorithms that have been used include reinforcement learning algorithms [33, 88] Qlearning [92, 109], classifier systems [41] and various forms of learning algorithms that have been adapted for use in automated markets [38, 91] Many of these learning algorithms were originally developed with optimality objectives in mind, so caution must be used in applying them to social processes. For ....

Tesauro, G. J., and Kephart, J. (1999). Pricing in agent economies using multi-agent Qlearning. In Proceedings of the Fifth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty , July.

....of the expected consumer response for any price pair, and moreover have full knowledge of both reward functions. Q learning is one of a variety of ways of endowing agents with foresight, i.e. an ability to antic ipate long term consequences of actions. Foresight was found in previous work (Tesauro and Kephart, 1999a,b) to improve profitability, and to damp out or eliminate the pathological behavior of unending cyclic price wars, in which long episodes of repeated undercutting amongst the sellers alternate with large jumps in price. Such price wars were found to be rampant in prior studies of agent economy ....

....that optimize immediate reward, but do not anticipate any longer term consequences. Q learning in particular is a principled way to obtain deep lookahead, since the Q function represents the cumulative discounted reward looking infinitely far ahead in time. In contrast, the prior work of (Tesauro and Kephart, 1999a) was based on shallow finite lookahead. Q learning with lookup tables was previously studied in (Tesauro and Kephart, 1999b) A single Qlearner playing against a myoptimal opponent always converged to an optimal policy. In the more interesting case of two agents simultaneously Q learning ....

[Article contains additional citation context not shown here]

G. Tesauro and J. O. Kephart,"Pricing in agent economies using multi-agent Q-learning." Proceedings of: Workshop on Decision Theoretic and Game Theoretic Agents, London, England, 5-6 July 1999(b).

.... informational and computational demands: game theoretic pricing (GT) myoptimal pricing (MY) derivative following (DF) and no regret learning (NR) Previously, we studied the dynamics that ensue when shopbot assisted buyers interact with pricebots utilizing only a subset of these strategies [19, 25, 29]. In this work, we simulate additional, more sophisticated, pricebot strategies, and nd that the game theoretic equilibrium can arise dynamically as the outcome of adaptive learning in our model of shopbots and pricebots. This paper is organized as follows. The next section presents our model of ....

G.J. Tesauro and J.O. Kephart. Pricing in agent economies using multi-agent q-learning. In Proceedings of Fifth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, pages 71-86, July 1999.

.... informational and computational demands: game theoretic pricing (GT) myoptimal pricing (MY) derivative following (DF) and no regret learning (NR) Previously, we studied the dynamics that ensue when shopbot assisted buyers interact with pricebots utilizing only a subset of these strategies [19, 25, 29]. In this work, we simulate additional, more sophisticated, pricebot strategies, and find that the game theoretic equilibrium can arise dynamically as the outcome of adaptive learning in our model of shopbots and pricebots. This paper is organized as follows. The next section presents our model ....

G.J. Tesauro and J.O. Kephart. Pricing in agent economies using multi-agent q-learning. In Proceedings of Fifth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, pages 71--86, July 1999.

.... and computational needs: game theoretic pricing (GT) myoptimal pricing (MY) derivative following (DF) and Q learning (Q) Previously, we studied the price and profit dynamics that ensue when shopbotassisted buyers interact with homogeneous collections of pricebots that utilize these algorithms [9, 10, 15]. In this work, we first establish that our previous results are not significantly altered when buyers valuations are inhomogeneous rather than identical. Later, we examine the behavior that ensues when pricebots employing different strategies are pitted against one another. This paper is ....

....stationary Markovian strategies. Situations that deviate from this, such as history dependent opponents or non stationary learning opponents (e.g. another Q learner) constitute an interesting and open research topic that is touched upon here and in some of our prior work (see Tesauro and Kephart [15]) 5 GT, MY, and DF Simulations We simulated an economy with 1000 buyers and initially 2, and later 5, pricebots employing various mixtures of pricing strategies. In the simulations depicted, buyer valuations are uniformly distributed in the interval [0; 1] and each seller s production cost r = ....

[Article contains additional citation context not shown here]

G. Tesauro and J. Kephart. Pricing in agent economies using multi-agent q-learning. In Proceedings of Fifth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, To Appear, July 1999.

No context found.

Tesauro, G., & Kephart, J. (1999). Pricing in agent economies using multi-agent Qlearning. Proceedings of Fifth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty (pp. 71--86).

No context found.

Tesauro, G., & Kephart, J. (1999). Pricing in agent economies using multi-agent Qlearning. Proceedings of Fifth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty (pp. 71--86).

No context found.

Tesauro, Gerald, and Jeffrey O. Kephart. 2002. "Pricing in Agent Economies Using Multi-Agent Q-Learning." Autonomous Agents and Multi-Agent Systems 5 eptem er): 289-304.

No context found.

G. Tesauro and J. O. Kephart. Pricing in agent economies using multi-agent Q-learning. Autonomous Agents and Multi-Agent Systems, 5(3):289-304, 2002.

No context found.

G. Tesauro and J. O. Kephart. Pricing in agent economies using multi-agent Q-learning. Autonomous Agents and Multi-Agent Systems, 5(3):289-304, 2002.

No context found.

Tesauro, G. and J. O. Kephart, #999. Pricing in agent economies using multi-agent Q-learning. Proceedings of Game Theoretic and Decision Theoretic Agents Workshop.

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