David Pardoe

Bidding for customer orders in TAC SCM

Supply chains are a current, challenging problem for agent-based electronic commerce. Motivated by the Trading Agent Competition Supply Chain Management (TAC SCM) scenario, we consider an individual supply chain agent as having three major subtasks: acquiring supplies, selling products, and managing its local manufacturing process. In this paper, we focus on the sales subtask. In particular, we consider the problem of finding the set of bids to customers in simultaneous reverse auctions that maximizes the agents expected profit. The key technical challenges we address are i) predicting the probability that a customer will accept a particular bid price, and ii) searching for the most profitable set of bids. We first compare several machine learning approaches to estimating the probability of bid acceptance. We then present a heuristic approach to searching for the optimal set of bids. Finally, we perform experiments in which we apply our learning method and bidding method during actual gameplay to measure the impact on agent performance.

TacTex-03: a supply chain management agent

This paper introduces TacTex-03, an agent designed to participate in the Trading Agent Competition Supply Chain Management Scenario (TAC SCM). As specified by this scenario, TacTex-03 acts as a simulated computer manufacturer in charge of buying components such as chips and motherboards, manufacturing different types of computers, and selling them to customers. TacTex-03 was the top scorer in two of the preliminary rounds of the 2003 TAC SCM competition, and finished in 3rd place in the finals.

Adaptive mechanism design: a metalearning approach

Auction mechanism design has traditionally been a largely analytic process, relying on assumptions such as fully rational bidders. In practice, however, bidders often exhibit unknown and variable behavior, making them difficult to model and complicating the design process. To address this challenge, we explore the use of an adaptive auction mechanism: one that learns to adjust its parameters in response to past empirical bidder behavior so as to maximize an objective function such as auctioneer revenue. In this paper, we give an overview of our general approach and then present an instantiation in a specific auction scenario. In addition, we show how predictions of possible bidder behavior can be incorporated into the adaptive mechanism through a metalearning process. The approach is fully implemented and tested. Results indicate that the adaptive mechanism is able to outperform any single fixed mechanism, and that the addition of metalearning improves performance substantially.

Developing adaptive auction mechanisms

Mechanism design has traditionally been a largely analytic process, relying on assumptions such as fully rational bidders. In practice, however, these assumptions may not hold, making bidder behavior difficult to model and complicating the design process. To address this issue, we propose a different approach to mechanism design. Instead of relying on analytic methods that require specific assumptions about bidders, our approach is to create a self-adapting mechanism that adjusts auction parameters in response to past auction results. In this paper, we describe our approach and then present an example of its implementation to illustrate its efficacy.

Evolving neural network ensembles for control problems

In neuroevolution, a genetic algorithm is used to evolve a neural network to perform a particular task. The standard approach is to evolve a population over a number of generations, and then select the final generations champion as the end result. However, it is possible that there is valuable information present in the population that is not captured by the champion. The standard approach ignores all such information. One possible solution to this problem is to combine multiple individuals from the final population into an ensemble. This approach has been successful in supervised classification tasks, and in this paper, it is extended to evolutionary reinforcement learning in control problems. The method is evaluated on a challenging extension of the classic pole balancing task, demonstrating that an ensemble can achieve significantly better performance than the champion alone.

The 2007 TAC SCM Prediction Challenge

The TAC SCM Prediction Challenge presents an opportunity for agents designed for the full TAC SCM game to compete solely on their ability to make predictions. Participants are presented with situations from actual TAC SCM games and are evaluated on their prediction accuracy in four categories: current and future computer prices, and current and future component prices. This paper introduces the Prediction Challenge and presents the results from 2007 along with an analysis of how the predictions of the participants compare to each other.

Adapting in agent-based markets: a study from TAC SCM

An agent attempting to model market conditions may benefit from considering how various combinations of competitor strategies would impact these conditions. We give an illustration using a prediction task faced by our agent for the Supply Chain Management scenario of the Trading Agent Competition (TAC SCM). We present the learning approach taken, evaluate its effectiveness, and then explore methods of improving predictions through combining multiple sources of data reflecting various combinations of competitor behaviors.

TacTex-05: an adaptive agent for TAC SCM

Supply chains are ubiquitous in the manufacturing of many complex products. Traditionally, supply chains have been created through the interactions of human representatives of the companies involved, but advances in autonomous agent technologies have sparked an interest in automating the process. The Trading Agent Competition Supply Chain Management (TAC SCM) scenario provides a unique testbed for studying supply chain management agents. This paper introduces TacTex-05, the champion agent from the 2005 competition, focusing on its ability to adapt to opponent behavior over a series of games. The impact of this adaptivity is examined through both analysis of competition results and controlled experiments.

Adapting price predictions in TAC SCM

In agent-based markets, adapting to the behavior of other agents is often necessary for success. When it is not possible to directly model individual competitors, an agent may instead model and adapt to the market conditions that result from competitor behavior. Such an agent could still benefit from reasoning about specific competitor strategies by considering how various combinations of these strategies would impact the conditions being modeled. We present an application of such an approach to a specific prediction problem faced by the agent TacTex-06 in the Trading Agent Competitions Supply Chain Management scenario (TAC SCM).

Agent-Based Supply Chain Management: Bidding for Customer Orders

Supply chains are a current, challenging problem for agent-based electronic commerce. Motivated by the Trading Agent Competition Supply Chain Management (TAC SCM) scenario, we consider an individual supply chain agent as having three major subtasks: acquiring supplies, selling products, and managing its local manufacturing process. In this paper, we focus on the sales subtask. In particular, we consider the problem of finding the set of bids to customers in simultaneous reverse auctions that maximizes the agentýs expected profit. The key technical challenges we address are i) predicting the probability that a customer will accept a particular bid price, and ii) searching for the most profitable set of bids. We first compare several machine learning approaches to estimating the probability of bid acceptance. We then present a heuristic approach to searching for the optimal set of bids. Finally, we perform experiments in which we apply our learning method and bidding method during actual gameplay to measure the impact on agent performance. Full details can be found in the extended version of this paper [1].