David Sarne

The Search for Coalition Formation in Costly Environments

We study the dynamics of forming coalitions of self-interested autonomous buyer agents, for the purpose of obtaining a volume discount. In our model, agents, representing coalitions of various sizes, may choose to be acquainted with other agents, hopefully ending up with a joint coalition structure, which will improve the overall price. Upon encountering potential partnering opportunities for extended coalitions, the agent needs to decide whether to accept or reject them. Each coalition partnership encapsulates expected benefit for the agent; however the process of finding a potential partner is associated with a cost. We explore the characteristics of the agent’s optimal strategies in the equilibrium and develop the equations from which these strategies can be derived. Efficient algorithms are suggested for a specific size-two variant of the problem, in order to demonstrate how each agent’s computation process can be significantly improved. These algorithms will be used as an infrastructure from which the general case algorithms can be extracted.

Multi-goal economic search using dynamic search structures

This paper investigates cooperative search strategies for agents engaged in costly search in a complex environment. Searching cooperatively, several search goals can be satisfied within a single search effort. Given the searchers’ preferences, the goal is to conduct a search in a way that the expected overall utility out of the set of opportunities found (e.g., products when operating in a market) minus the costs associated with finding that set is maximized. This search scheme, given in the context of a group search, applies also to scenarios where a single agent has to search for a set of items for satisfying several different goals. The uniqueness of the proposed mechanism is in the ability to partition the group of agents/goals into sub-groups where the search continues for each group autonomously. As we show throughout the paper, this strategy is favorable as it weakly dominates (i.e., can improve but never worsen) cooperative and autonomous search techniques. The paper presents a comprehensive analysis of the new search method and highlights the specific characteristics of the optimal search strategy. Furthermore, we introduce innovative algorithms for extracting the optimal search strategy in a range of common environments, that eliminates the computational overhead associated with the use of the partitioning technique.

Evaluating the Applicability of Peer-Designed Agents in Mechanisms Evaluation

In this paper we empirically investigate the feasibility of using peer-designed agents (PDAs) instead of people for the purpose of mechanism evaluation. This latter approach has been increasingly advocated in agent research in recent years, mainly due to its many benefits in terms of time and cost. Our experiments compare the behavior of 31 PDAs and 150 people in a legacy eCommerce-based price-exploration setting, using different price-setting mechanisms and different performance measures. The results show a varying level of similarity between the aggregate behavior obtained when using people and when using PDAs -- in some settings similar results were obtained, in others the use of PDAs rather than people yields substantial differences. This suggests that the ability to generalize results from one successful implementation of PDA-based systems to another, regarding the use of PDAs as a substitute to people in systems evaluation, is quite limited. The decision to prefer PDAs for mechanism evaluation is therefore setting dependent and the applicability of the approach must be re-evaluated whenever switching to a new setting or using a different measure. Furthermore, we show that even in settings where the aggregate behavior is found to be similar, the individual strategies used by agents in each group highly vary.

Expert-mediated sequential search

This paper studies markets, such as Internet marketplaces for used cars or mortgages, in which consumers engage in sequential search. In particular, we consider the impact of information-brokers (experts) who can, for a fee, provide better information on true values of opportunities. We characterize the optimal search strategy given a price and the terms of service set by the expert, and show how to use this characterization to solve the monopolist expert’s service pricing problem. Our analysis enables the investigation of three common pricing schemes (pay-per-use, unlimited subscription, and package pricing) that can be used by the expert. We demonstrate that in settings characteristic of electronic marketplaces, namely those with lower search costs for consumers and lower costs of production of expert services, unlimited subscription schemes are favored. Finally, we show that the platform that connects consumer and experts can improve social welfare by subsidizing the purchase of expert services. The optimal level of subsidy forces the buyer to exactly fully internalize the marginal cost of provision of expert services. In electronic markets, this cost is minimal, so it may be worthwhile for the platform to make the expert freely available to consumers.

Physical search problems with probabilistic knowledge

This paper considers the problem of an agent or a team of agents searching for a resource or tangible good in a physical environment, where the resource or good may possibly be obtained at one of several locations. The cost of acquiring the resource or good at a given location is uncertain (a priori), and the agents can observe the true cost only when physically arriving at this location. Sample applications include agents in exploration and patrol missions (e.g., an agent seeking to find the best location to deploy sensing equipment along its path). The uniqueness of these settings is in that the cost of observing a new location is determined by distance from the current one, impacting the consideration for the optimal search order. Although this model captures many real world scenarios, it has not been investigated so far. We analyze three variants of the problem, differing in their objective: minimizing the total expected cost, maximizing the success probability given an initial budget, and minimizing the budget necessary to obtain a given success probability. For each variant, we first introduce and analyze the problem with a single agent, either providing a polynomial solution to the problem or proving it is NP-complete. We also introduce a fully polynomial time approximation scheme algorithm for the minimum budget variant. In the multi-agent case, we analyze two models for managing resources, shared and private budget models. We present polynomial algorithms that work for any fixed number of agents, in the shared or private budget model. For non-communicating agents in the private budget model, we present a polynomial algorithm that is suitable for any number of agents. We also analyze the difference between homogeneous and heterogeneous agents, both with respect to their allotted resources and with respect to their capabilities. Finally, we define our problem in an environment with self-interested agents. We show how to find a Nash equilibrium in polynomial time, and prove that the bound on the performance of our algorithms, with respect to the social welfare, is tight.

Problem restructuring for better decision making in recurring decision situations

This paper proposes the use of restructuring information about choices to improve the performance of computer agents on recurring sequentially dependent decisions. The intended situations of use for the restructuring methods it defines are website platforms such as electronic marketplaces in which agents typically engage in sequentially dependent decisions. With the proposed methods, such platforms can improve agents’ experience, thus attracting more customers to their sites. In sequentially-dependent-decisions settings, decisions made at one time may affect decisions made later; hence, the best choice at any point depends not only on the options at that point, but also on future conditions and the decisions made in them. This “problem restructuring” approach was tested on sequential economic search, which is a common type of recurring sequentially dependent decision-making problem that arises in a broad range of areas. The paper introduces four heuristics for restructuring the choices that are available to decision makers in economic search applications. Three of these heuristics are based on characteristics of the choices, not of the decision maker. The fourth heuristic requires information about a decision-makers prior decision-making, which it uses to classify the decision-maker. The classification type is used to choose the best of the three other heuristics. The heuristics were extensively tested on a large number of agents designed by different people with skills similar to those of a typical agent developer. The results demonstrate that the problem-restructuring approach is a promising one for improving the performance of agents on sequentially dependent decisions. Although there was a minor degradation in performance for a small portion of the agents, the overall and average individual performance improved substantially. Complementary experimentation with people demonstrated that the methods carry over, to some extent, also to human decision makers. Interestingly, the heuristic that adapts based on a decision-maker’s history achieved the best results for computer agents, but not for people.

Competitive Shopbots-Mediated Markets

This article considers markets mediated by autonomous self-interested comparison-shopping agents. As in today’s markets, the agents do not charge buyers for their services but rather benefit from payments obtained from sellers upon the execution of a transaction. The agents aim at maximizing their expected benefit, taking into consideration the cost incurred by the search and competition dynamics that arise in the multi-agent setting. This article provides a comprehensive analysis of such models, based on search theory principles. The analysis results in a characterization of the buyers’ and agents’ search strategies in equilibrium. The main result of this article is that the use of self-interested comparison-shopping agents can result in a beneficial equilibrium, where both buyers and sellers benefit, in comparison to the case where buyers control the comparison-shopping agent, and the comparison-shopping agents necessarily do not lose. This, despite the fact that the service is offered for free to buyers and its cost is essentially covered by sellers. The analysis generalizes to any setting where buyers can use self-interested agents capable of effectively performing the search (e.g., evaluating opportunities) on their behalf.

Time-Variant Distributed Agent Matching Applications

The process of pair partnership formation is an important infrastructure for many plausible MAS applications. Each agent evaluates potential partner agents, where each potential match yields a different utility. Commonly, the utility associated with a given agent partner in such two-sided search processes may change over time. This change in the agentýs future attractiveness to potential partners significantly increases the complexity of the agentýs decision making process regarding the set of agents it is willing to partner with. In this paper we analyze the special dynamics and present equilibrium characteristics of such a model. The agents can gain a utility derived from the partner agentýs type. However, as an agent has an incentive to extend its search for a better type partner, the benefit that can be offered to potential partners reduces as the search proceeds. We introduce a two-sided model which takes into consideration a continuous decrease in the agentýs type and formulate the appropriate equilibrium equations. The suggested equilibrium analysis yields an algorithm for the calculation of the agentsý equilibrium strategy. Special emphasis is placed on the scenario where an agentýs attractiveness is influenced by an additional dimension other than just time. Simulation results are presented to illustrate the findings.

Determining the value of information for collaborative multi-agent planning

This paper addresses the problem of computing the value of information in settings in which the people using an autonomous-agent system have access to information not directly available to the system itself. To know whether to interrupt a user for this information, the agent needs to determine its value. The fact that the agent typically does not know the exact information the user has and so must evaluate several alternative possibilities significantly increases the complexity of the value-of-information calculation. The paper addresses this problem as it arises in multi-agent task planning and scheduling with architectures in which information about the task schedule resides in a separate “scheduler” module. For such systems, calculating the value to overall agent performance of potential new information requires that the system component that interacts with the user query the scheduler. The cost of this querying and inter-module communication itself substantially affects system performance and must be taken into account. The paper provides a decision-theoretic algorithm for determining the value of information the system might acquire, query-reduction methods that decrease the number of queries the algorithm makes to the scheduler, and methods for ordering the queries to enable faster decision-making. These methods were evaluated in the context of a collaborative interface for an automated scheduling agent. Experimental results demonstrate the significant decrease achieved by using the query-reduction methods in the number of queries needed for reasoning about the value of information. They also show the ordering methods substantially increase the rate of value accumulation, enabling faster determination of whether to interrupt the user.

Strategic information platforms: selective disclosure and the price of "free"

This paper deals with platforms that provide agents easier access to the type of opportunities in which they are interested (e.g., eCommerce platforms, used cars bulletins and dating web-sites). We show that under various common service schemes, a platform can benefit from not necessarily listing all the opportunities with which it is familiar, even if there is no marginal cost for listing any additional opportunity. The main implication of this result is that platforms should extract their expected-profit-maximizing service terms not based solely on the fees charged from users, but they should also use the subset that will be listed as the decision variable in the optimization problem. The analysis applies to four well-known service schemes that a platform may use to price its services. We show that neither of these schemes generally dominates the others or is dominated by any of the others. For the common case of homogeneous preferences, however, several dominance relationships can be proved, enabling the platform to identify the schemes that should be used as a default. Furthermore, the analysis provides a game-theoretic search-based explanation for a possible preference of buyers to pay for the service rather than receive it for free (e.g., when the service is sponsored by ads), a phenomena that has been justified in prior literature typically with the argument of willingness to pay a premium for an ad-free experience or more reliable platforms. The paper shows that this preference can hold both for the users and the platform in a given setting, even if both sides are fully strategic.