Browsing by Author "Gini, Maria"

Now showing 1 - 20 of 31

A Market Architecture for Multi-Agent Contracting
(1997) Collins, John; Jamison, Scott; Mobasher, Bamshad; Gini, Maria
We present a generalized market architecture that provides support for a variety of types of transactions, from simple buying and selling of goods and services to complex multi-agent contract negotiations. This architecture is organized around three basic components: the exchange, the market, and the session. We also present a negotiation protocol for planning by contracting that takes advantage of the services of the market. We show how the existence of an appropriate market infrastructure can add value to a multi-agent contracting protocol by controlling fraud and discouraging counterspeculation.
A Minimally Constrained Environment for the Study of Cooperation
(2008-04-24) Damer, Steven; Gini, Maria
We describe a simple environment to study cooperation between two agents and a method of achieving cooperation in that environment. The environment consists of randomly generated normal form games with uniformly distributed payoffs. Agents play multiple games against each other, each game drawn independently from the random distribution. This environment provides a good model of the difficulties of cooperating in an ever changing world. Tit-for-Tat cannot be used because moves are not labeled as "cooperate" or "defect", fictitious play cannot be used because the agent never sees the same game twice, and approaches suitable for stochastic games cannot be used because the set of states is not finite. Our agent identifies cooperative moves by assigning an attitude to its opponent and to itself. The attitude determines how much a player values its opponents payoff, i.e how much the player is willing to deviate from strictly self-interested behavior. To cooperate, our agent estimates the attitude of its opponent by observing its moves and reciprocates by setting its own attitude accordingly. We show how the opponent's attitude can be estimated using a particle filter, even when the opponent is changing its attitude.
A regression model for predicting optimal purchase timing for airline tickets
(2011-10-18) Groves, William; Gini, Maria
Optimal timing for airline ticket purchasing from the consumer's perspective is challenging principally because buyers have insufficient information for reasoning about future price movements. This paper presents a model for computing expected future prices and reasoning about the risk of price changes. The proposed model is used to predict the future expected minimum price of all available flights on specific routes and dates based on a corpus of historical price quotes. Also, we apply our model to predict prices of flights with specific desirable properties such as flights from a specific airline, non-stop only flights, or multi-segment flights. By comparing models with different target properties, buyers can determine the likely cost of their preferences. We present the expected costs of various preferences for two high-volume routes. Performance of the prediction models presented is achieved by including instances of time-delayed features, by imposing a class hierarchy among the raw features based on feature similarity, and by pruning the classes of features used in prediction based on in-situ performance. Our results show that purchase policy guidance using these models can lower the average cost of purchases in the 2 month period prior to a desired departure. The proposed method compares favorably with a deployed commercial web site providing similar purchase policy recommendations.
A Taxonomy for Task Allocation Problems with Temporal and Ordering Constraints
(2016-05-06) Nunes, Ernesto; Manner, Marie; Mitiche, Hakim; Gini, Maria
Previous work on assigning tasks to robots has proposed extensive categorizations of allocation of tasks with and without constraints. The main contribution of this paper is a more specific categorization of problems that have both temporal and ordering constraints. We propose a novel taxonomy that builds on the existing taxonomy for multi-robot task allocation and organizes the current literature according to the temporal nature of the tasks. We summarize widely used models and methods from the task allocation literature and related areas, such as vehicle routing and scheduling problems, showing similarities and differences.
An Overview of XRobots: A Hierarchical State Machine-Based Language
(2011) Tousignant, Steve; Van Wyk, Eric; Gini, Maria
This paper introduces a prototype domain-specific language for programming mobile robots that is based on hierarchical state machines. A novelty of this language is that states are treated as first class entities in the language and thus they can be passed as arguments to other parameterized states. The structure and behavior of the language is presented, along with an example program. Further work and language design challenges are also discussed.
Artificial Emotional Intelligence: Dialogue Systems in Medicine
(2019) Arun, Vishnu; Huffstutler, Thomas; Finzel, Raymond; Ferland, Libby; Pakhomov, Serguei; Gini, Maria
Asking the Right Question: Risk and Expectation in Multi-Agent Contracting
(2002-11-05) Babanov, Alexander; Collins, John; Gini, Maria
This paper investigates methods of reducing risk in market-based auctions of tasks with complex time constraints and interdependencies. The research addresses problems in a contracting setting in which a buyer has a set of tasks to be performed. Because of the complex dependencies among the tasks, a task not completed on time might have devastating effect on other tasks. Therefore, the problem is to sequence tasks and allocate time windows to maximize the expected utility of the agent. Because there is a probability of loss as well as a probability of gain, the decision process must deal with the risk posture of the person or organization on whose behalf the decision maker is acting.
An Assessment of the Writing of Undergraduate Computer Science Students
(University of Minnesota, 2002) Nurkkala, Tom; Gini, Maria
Automated Route Planning and Optimizing Software
(Minnesota Department of Transportation, 1997-02) Gini, Maria; Zhao, Yiyuan
This report presents the results of a study of automated route planning and optimizing software to be used by Mn/DOT Metro Division and by Hennepin County for snow plow and for snow and ice control logistical planning. The study has: 1. Produced a uniform set of specifications for the two agencies; 2. Identifyed and analyzed a large number of commercially available software simulation packages for route and logistical planning; 3. Prepared recommendations on how to proceed with the project with a detailed analysis of their advantages and shortcomings.
Automation for Regulated Issue Tracking Activities
(2012-03-26) Drew, Touby; Gini, Maria
We describe the application of automated support for issue tracking and related software engineering activities of development teams at the world's largest medical device manufacturer. We present some challenges and classes of defects found in product software, related artifacts, and the issues which track them. We describe enhanced means for defect detection, data mining and analysis, and other novel support we provide at the time of issue review. Finally, we describe evidence of the positive impact of this support, its adoption, lessons learned and potential next steps.
Component-based design for a trading agent
(2004-05-04) Collins, John; Agovic, Amrudin; Damer, Steven; Gini, Maria
MinneTAC is an agent designed to compete in the Supply-Chain Trading Agent Competition. It is also designed to support the needs of a group of researchers, each of whom is interested in different decision problems related to the competition scenario. The design of MinneTAC breaks out each basic behavior into a separate, configurable component. Dependencies between components are almost non-existent. The agent itself is defined as a set of "roles", and a working agent is one for which a component is supplied for each role. This allows each user to focus on a single problem and work independently, and it allows multiple user to tackle the same problem in different ways. A working MinneTAC agent is completely defined by a set of configuration files that map the desired roles to the code that implements them, and that set parameters for the components. This paper describes the design of MinneTAC and evaluates its effectiveness in support of our research agenda, and in its competitiveness in the TAC-SCM game environment.
Design and Analysis of the MinneTAC-03 Supply-Chain Trading Agent
(2004-04-26) Ketter, Wolfgang; Kryzhnyaya, Elena; Damer, Steven; McMillen, Colin; Agovic, Amrudin; Collins, John; Gini, Maria
MinneTAC is an agent designed to compete in the Supply-Chain Trading Agent Competition. It is also designed to support the needs of a group of researchers, each of whom is interested in different decision problems related to the competition scenario. The design of MinneTAC breaks outeach basic behavior into a separate, configurable component. Dependencies between components are almost non-existent. This design allows each user to focus on a single problem and work independently, and it allows multiple user to tackle the same problem in different ways. This paper describes the design of MinneTAC and evaluates its effectiveness in support of our research agenda, and in its competitiveness in the TAC-SCM game environment. We also describe two sales strategies used by MinneTAC. Both strategies estimate, as the game progresses, the probability of receiving a customer order for different prices and compute the expected profit. Offers are made to maximize the expected profit on each order. The maindifference between the two strategies is in how the probability of receiving an order and the offer prices are computed. The first strategy works well in high-demand games, the second was developed to improve performance in low-demand games. We empirically analyze the effect of the discount given by suppliers on orders received the firstday of the game, and we show that in high-demand games there is a strong correlation between the offers an agent receives from suppliers on the first day of the game and the agent's performance in the game.
Detecting and Forecasting Economic Regimes in Automated Exchanges
(2007-03-05) Ketter, Wolfgang; Collins, John; Gini, Maria; Gupta, Alok; Schrater, Paul
We present basic building blocks of an agent that can use observable market conditions to characterize the microeconomic conditions of the market and predict future market trends. The agent can use this information to make both tactical decisions such as pricing and strategic decisions such as product mix and production planning. We develop methods that can learn dominant market conditions, such as over-supply or scarcity, from historical data using computational methods to construct price density functions. We discuss how this knowledge can be used, together with real-time observable information, to identify the current dominant market condition and to forecast market changes over a planning horizon. We validate our methods by presenting experimental results in a case study, the Trading Agent Competition for Supply Chain Management.
Efficient Statistical Methods for Evaluating Trading Agent Performance
(2007-02-13) Sodomka, Eric; Collins, John; Gini, Maria
Market simulations, like their real-world counterparts, are typically domains of high complexity, high variability, and incomplete information. The performance of autonomous agents in these markets depends both upon the strategies of their opponents and on various market conditions, such as supply and demand. Because the space for possible strategies and market conditions is very large, empirical analysis in these domains becomes exceedingly difficult. Researchers who wish to evaluate their agents must run many test games across multiple opponent sets and market conditions to verify that agent performance has actually improved. Our approach is to improve the statistical power of market simulation experiments by controlling their complexity, thereby creating an environment more conducive to structured agent testing and analysis. We develop a tool that controls variability across games in one such market environment, the Trading Agent Competition for Supply Chain Management (TAC SCM), and demonstrate how it provides an efficient, systematic method for TAC SCM researchers to analyze agent performance.
Episode 123: Artificial Intelligence in Education
(2023-07-28) Sirovy, Kaylie; Gini, Maria; Chancellor, Stevie
The Minnesota Daily sat down with CSE professors to explore the rising trend of ChatGPTand its implications in the field of education.
Evaluating Risk: Flexibility and Feasibility in Multi-Agent Contracting
(1999-01-09) Sundareswara, Rashmi; Tsvetovat, Maksim; Collins, John; Gini, Maria; van Tonder, Joshua; Mobasher, Bamshad
In an automated contracting environment, where a "customer" agent must negotiate with other self-interested "supplier" agents in order to execute its plans, there is a trade-off between giving the suppliers sufficient flexibility to incorporate the requirements of the customer's call-for-bids into their own resource schedules, and ensuring the customer that any bids received can be composed into a feasible plan. In this paper, we introduce a bid evaluation process that incorporates cost, task coverage, temporal feasibility, and risk estimation. Using this evaluation process, we provide an empirical study of the trade-offs between flexibility, plan feasibility, and cost in the context of our MAGNET multi-agent contracting market infrastructure. Our experimental results demonstrate that the advantage of increasing supplier flexibilty is dependent on the number of available suppliers. In other words, if the number of suppliers is small, the risk of plan nfeasibility outweighs the advantage of added flexibility. On the other hand, if the numer of suppliers is large, the more flexibile plan specifications result in lower-risk plans.
Exploring Decision Processes in Multi-Agent Automated Contracting
(2000-10-16) Collins, John; Gini, Maria
We are interested in the problem of multi-agent contracting, in which customers must solicit the resources and capabilities of other, self-interested agents in order to accomplish their goals. Goals may involve the execution of multi-step plans, in which different steps are contracted out to different suppliers. We have focused on decision criteria for composing requests for quotations, managing the bidding process, evaluating bids, and monitoring plan execution. We have developed a testbed that allows us to study these decision behaviors. It can generate sets of plans with known statistical attributes, formulate and submit requests for quotations, generate bids with well-defined statistics, and evaluate those bids according to a number of criteria. Each of these processes is supported by an abstract interface and a series of pluggable modules with a large number of configuration parameters. Data collection and analysis tools round out the package. We will demonstrate how to take statistics from a real application domain, apply them to the simulation, and test a variety of bid-management and bid-evaluation procedures against them.
Flexible decision support in dynamic interorganizational networks
(2009-11-23) Collins, John; Ketter, Wolfgang; Gini, Maria
An effective Decision Support System (DSS) should help its users improve decision-making in complex, information-rich, dynamic environments. We present a feature gap analysis of current decision support technologies, and we identify a set of DSS Desiderata, properties that can contribute both effectiveness and flexibility to users in such environments. We show that there is a gap between the features provided by current DSS technologies and the DSS Desiderata we aim for. We present a design-science approach that extends the boundaries of human decision-makers by creating a new and innovative artifact called "evaluator service networks" at the confluence of people, organizations, and technology. Our artifact enables users to compose decision behaviors from separate, configurable components, and allows dynamic construction of analysis and modeling tools from small, single-purpose evaluator services. The result is a network that can easily be configured to test hypotheses and analyze the impact of various choices for elements of decision processes. We have implemented and tested this design in an interactive version of the MinneTAC trading agent, an agent designed for the Trading Agent Competition for Supply Chain Management. We present an example of an evaluator service network that determines sales prices in a rich, dynamic trading environment. Additionally we describe visual interface elements that allow users to see and manipulate the configuration of the network, and to construct economic dashboards that can display the current and historical state of any node in the network.
Flexible decision support in dynamic interorganizational networks
(2008-11-10) Collins, John; Ketter, Wolfgang; Gini, Maria
This Technical Report has been completely re-written and re-released as #09-028. The updated version is available for download at its new location: http://www.cs.umn.edu/research/technical_reports.php?page=report&report_id=09-028
Forming Long Term Teams to Exploit Synergies among Heterogeneous Agents
(2012-05-17) Parker, James; Nunes, Ernesto; Godoy, Julio; Gini, Maria
In this work, we describe how agents can take advantage of synergies among each other and increase their efficiency at accomplishing tasks by working in teams. We present different team structures for heterogeneous agents, which range from static teams that are formed upfront and never change, to dynamic teams where agents can change teams as need arises, to teams that allow only some types of agents to be members. We also develop teaming strategies that are strongly domain specific for the RoboCup Rescue simulation environment. RoboCup Rescue is a simulated environment which is characterized by uncertainty in available information and by severely limited communications among agents. The tasks to be accomplished are saving civilians who are trapped in buildings and preventing fires from spreading in the city. The locations of civilians and fires are not known upfront and have to be discovered. The domain constraints limit the applicability of some popular team formation algorithms and require adaptive strategies. We measure the effectiveness of the various teaming strategies we propose. Our experimental results support the hypothesis that teaming improves performance, and that more specialized and knowledge rich teaming arrangements perform better.

University Digital Conservancy

Browse by Author

Browsing by Author "Gini, Maria"