Probabilistic Automated Bidding in Multiple Auctions

Dumas, Marlon; Aldred, Lachlan; Governatori, Guido; Hofstede, Arthur H. M. ter

doi:10.1023/b:elec.0000045972.75533.90

Cited by 28 publications

(12 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The plot, which enables easier tracking of the learning process, presents the results of the all-pay auction. 17 The left side of the plot shows a noticeable preliminary advantage of the DVRL method. Particularly, the payoff of DVRL was higher than the others until the ninth round, where it became approximately equal to DVG.…”

Section: Discussion Of the Resultsmentioning

confidence: 97%

“…In many auctions, substitutability or complementarity exists between two or more items for sale [6,17,22,38]. This situation is especially common in web-based auctions, where one can find many simultaneous auctions of similar popular items on different sites [7,51].…”

Section: Simultaneous Cliff-edge Environmentsmentioning

confidence: 99%

“…Dumas et al [17] developed a probabilistic bidding agent that participates in multiple auctions. The agent gathers the bidding histories for each relevant auction house in order to predict the probability of each bid to win.…”

Section: Previous Approachesmentioning

confidence: 99%

See 2 more Smart Citations

Efficient bidding strategies for Cliff-Edge problems

Azoulay

Katz

Kraus

2013

Auton Agent Multi-Agent Syst

View full text Add to dashboard Cite

In this paper, we propose an efficient agent for competing in Cliff-Edge (CE) and simultaneous Cliff-Edge (SCE) situations. In CE interactions, which include common interactions such as sealed-bid auctions, dynamic pricing and the ultimatum game (UG), the probability of success decreases monotonically as the reward for success increases. This trade-off exists also in SCE interactions, which include simultaneous auctions and various multi-player ultimatum games, where the agent has to decide about more than one offer or bid simultaneously. Our agent competes repeatedly in one-shot interactions, each time against different human opponents. The agent learns the general pattern of the population's behavior, and its performance is evaluated based on all of the interactions in which it participates. We propose a generic approach which may help the agent compete against unknown opponents in different environments where CE and SCE interactions exist, where the agent has a relatively large number of alternatives and where its achievements in the first several dozen interactions are important. The underlying mechanism we propose for CE interactions is a new metaalgorithm, deviated virtual learning (DVL), which extends existing methods to efficiently cope with environments comprising a large number of alternative decisions at each decision point. Another competitive approach is the Bayesian approach, which learns the opponents' statistical distribution, given prior knowledge about the type of distribution. For the SCE, we propose the simultaneous deviated virtual reinforcement learning algorithm (SDVRL), the segmentation meta-algorithm as a method for extending different basic algorithms, and a heuristic called fixed success probabilities (FSP). Experiments comparing the performance of the proposed algorithms with algorithms taken from the literature, as well as other intuitive meta-algorithms, reveal superiority of the proposed algorithms in average payoff and stability as well as in accuracy in converging to the optimal action, both in CE and SCE problems.

show abstract

Section: Discussion Of the Resultsmentioning

confidence: 97%

Section: Simultaneous Cliff-edge Environmentsmentioning

confidence: 99%

See 1 more Smart Citation

Efficient bidding strategies for Cliff-Edge problems

Azoulay

Katz

Kraus

2013

Auton Agent Multi-Agent Syst

View full text Add to dashboard Cite

show abstract

“…Given the selection of NCDE methods at the end of previous section, we are ready to apply reframing to several kinds of cost-sensitive decision problems, where particular families of loss functions are used. For instance, in this section we will explore a family of problems which are very common in econometrics, commerce and retailing applications, where we need to estimate the price (or other quantifiable features) for an offer or bid in the context of a sale, deal or auction [Schapire et al 2002;Dumas et al 2005;Kitts and Leblanc 2004;Wellman et al 2004;Ghani 2005]. One of the most relevant features of the loss functions in these applications is that they are highly discontinuous, since an offer which is much too expensive changes loss dramatically: from the maximum attainable benefit to no benefit at all (the offer is not accepted).…”

Section: Bid Applicationsmentioning

confidence: 99%

Probabilistic Reframing for Cost-Sensitive Regression

́ndez-Orallo

2014

ACM Trans. Knowl. Discov. Data

View full text Add to dashboard Cite

Common day applications of predictive models usually involve a full use of the available contextual information. When the operating context changes, one may fine-tune the by-default (incontextual) prediction or may even abstain from predicting a value (a reject). Global reframing solutions, where the same function is applied to adapt the estimated outputs to a new cost context, is a possible solution here. An alternative approach, which has not been studied in a comprehensive way for regression in the knowledge discovery and data mining literature, is the use of a local (e.g., probabilistic) reframing approach, where decisions are made according to the estimated output and a reliability, confidence or probability estimation. In this paper, we advocate for a simple two-parameter (mean and variance) approach, working with a normal conditional probability density. Given the conditional mean produced by any regression technique, we develop lightweight 'enrichment' methods that produce good estimates of the conditional variance, which are used by the probabilistic (local) reframing methods. We apply these methods to some very common families of cost-sensitive problems, such as optimal predictions in (auction) bids, asymmetric loss scenarios and rejection rules.

show abstract

“…The work in [9] presents a decision theoretic framework that an autonomous agent can use to bid effectively across multiple, simultaneous auctions. An approach to develop bidding agents that participate in multiple alternative auctions, with the goal of obtaining an item with a given probability is demonstrated in [10]. The authors in [11] develop a bidding agent for multiple heterogeneous auctions and build a heuristic decision making framework that an autonomous agent can exploit to tackle the problem of bidding across multiple auctions with varying start and end times and with varying protocols.…”

Section: Introductionmentioning

confidence: 99%

Automated Bidding in English Auctions Using Cognitive Agents in Mobile e-Commerce

Sidnal¹,

Manvi²

2012

IJIET

View full text Add to dashboard Cite

Abstract-There is an increasing demand to monitor and bid concurrently in multiple English auctions in order to procure the best deal for the desired products. This paper proposes a scheme to incorporate human intelligence and perspective (cognitive) in agents for bidding in multiple concurrent English auctions in mobile E-commerce. The bidder mobile agent is embedded with intelligence to take decisions on behalf of a bidder with regard to either continue bidding or withdraw from auctions. The agent ensures that the bidder receives an appropriate number of desired products from one of the participated relevant auction site that has better chances of winning with satisfaction.Dynamic E-market and mobile bidder scenarios are simulated to train the cognitive agents for enabling them to take decisions on behalf of the bidder in real time. The proposed scheme performs better than the existing bidding strategy (based on limited E-market parameters) in terms of best possible bid value, bidder satisfaction, and budget.

show abstract

Probabilistic Automated Bidding in Multiple Auctions

Cited by 28 publications

References 17 publications

Efficient bidding strategies for Cliff-Edge problems

Efficient bidding strategies for Cliff-Edge problems

Probabilistic Reframing for Cost-Sensitive Regression

Automated Bidding in English Auctions Using Cognitive Agents in Mobile e-Commerce

Contact Info

Product

Resources

About