Maria Gini

JAAMAS Journal 2026 Journal Article

Guest Introduction

• Maria Gini
• Jeffrey S. Rosenschein

AAMAS Conference 2023 Conference Paper

Provably Manipulable 3D Structures using Graph Theory

• John Harwell
• London Lowmanstone
• Maria Gini

We identify barriers to a broader application of multi-robot systems to construction and deconstruction tasks, which represent important real-world problems, such as repairing critical infrastructure of roads and levies after a disaster. We frame these tasks as instances of the parallel bricklayer problem, where independent agents must coordinate to concurrently manipulate aspects of a 3D environment without deadlocks. We extract desirable properties of graphs representing natural 3D structures and sketch a graphical representation to model and reason about structures composed of discrete cuboid blocks. We present a sample algorithm sketch for a non-trivial structure utilizing our model.

AAMAS Conference 2022 Conference Paper

SIERRA: A Modular Framework for Research Automation

• John Harwell
• London Lowmanstone
• Maria Gini

Modern intelligent systems researchers form hypotheses about system behavior and then run experiments using one or more independent variables to test their hypotheses. We present SIERRA, a novel framework structured around that idea for accelerating research developments and improving reproducibility of results. SIERRA makes it easy to quickly specify the independent variable(s) for an experiment, generate experimental inputs, automatically run the experiment, and process the results to generate deliverables such as graphs and videos. SIERRA provides reproducible automation independent of the execution environment (HPC hardware, real robots, etc.) and targeted platform (arbitrary simulator or real robots), enabling exact experiment replication (up to the limit of the execution environment and platform). It employs a deeply modular approach that allows easy customization and extension of automation for the needs of individual researchers, thereby eliminating manual experiment configuration and result processing via throw-away scripts.

IJCAI Conference 2019 Conference Paper

Swarm Engineering Through Quantitative Measurement of Swarm Robotic Principles in a 10, 000 Robot Swarm

• John Harwell
• Maria Gini

When designing swarm-robotic systems, system- atic comparison of algorithms from different do- mains is necessary to determine which is capa- ble of scaling up to handle the target problem size and target operating conditions. We propose a set of quantitative metrics for scalability, flexibility, and emergence which are capable of addressing these needs during the system design process. We demonstrate the applicability of our proposed met- rics as a design tool by solving a large object gath- ering problem in temporally varying operating con- ditions using iterative hypothesis evaluation. We provide experimental results obtained in simulation for swarms of over 10, 000 robots.

AAMAS Conference 2019 Conference Paper

The Gift Exchange Game: Managing Opponent Actions

• Steven Damer
• Maria Gini
• Jeffrey S. Rosenschein

Interacting with an opponent is a fundamental concern in multiagent systems. In this work, we consider ways in which an agent can manipulate an opponent to adopt a preferred strategy. This difficult problem is often further complicated by the difficulty of analyzing the game. We have developed the Gift Exchange game, a sequential game that is deliberately simplified to focus on how to interact with an opponent. In this paper we describe the game and discuss different methods an agent might use to influence its opponent to select a preferred action. We show results from using simulated annealing to find optimal strategies to use against a learning opponent.

AAAI Conference 2018 Conference Paper

Model-Free Iterative Temporal Appliance Discovery for Unsupervised Electricity Disaggregation

• Mark Valovage
• Akshay Shekhawat
• Maria Gini

Electricity disaggregation identifies individual appliances from one or more aggregate data streams and has immense potential to reduce residential and commercial electrical waste. Since supervised learning methods rely on meticulously labeled training samples that are expensive to obtain, unsupervised methods show the most promise for widespread application. However, unsupervised learning methods previously applied to electricity disaggregation suffer from critical limitations. This paper introduces the concept of iterative appliance discovery, a novel unsupervised disaggregation method that progressively identifies the ‘easiest to find’ or ‘most likely’ appliances first. Once these simpler appliances have been identified, the computational complexity of the search space can be significantly reduced, enabling iterative discovery to identify more complex appliances. We test iterative appliance discovery against an existing competitive unsupervised method using two publicly available datasets. Results using different sampling rates show iterative discovery has faster runtimes and produces better accuracy. Furthermore, iterative discovery does not require prior knowledge of appliance characteristics and demonstrates unprecedented scalability to identify long, overlapped sequences that other unsupervised learning algorithms cannot.

AAMAS Conference 2018 Conference Paper

Online Multi-Robot Coverage: Algorithm Comparisons

• Elizabeth A. Jensen
• Maria Gini

We consider the common assumptions made when multi-robot systems are used for exploration and coverage and the metrics used to compare performance. We then take three algorithms – the Rolling Dispersion Algorithm (RDA), the Multi-Robot Depth-First- Search (MR-DFS) algorithm, and the BoB algorithm – chosen for their different strengths and assumptions, and compare, using a set of common metrics, their performance in different simulation environments. We present two simple extensions to RDA – RDA- MS (multi-start) and RDA-EC (extended communication), which preserve RDA’s original assumptions, but are able to perform as well as the algorithms that make more demanding assumptions.

AAMAS Conference 2017 Conference Paper

Label Correction and Event Detection for Electricity Disaggregation

• Mark Valovage
• Maria Gini

Electricity disaggregation focuses on identifying individual appliances from one or more aggregate signals. By reporting detailed appliance usage to consumers, disaggregation has the potential to significantly reduce electrical waste in residential and commercial sectors. However, application of existing methods is limited by two critical shortcomings. First, supervised learning methods implicitly assume errorfree labels in training data, an unrealistic expectation for imperfectly-labeled consumer data. Second, supervised and unsupervised learning methods require parameters to be tuned to individual appliances and/or datasets, limiting widespread application. To address these limitations, this paper introduces the implementation of Bayesian changepoint detection (BCD) with necessary adaptations to electricity disaggregation. We introduce an algorithm to effectively apply BCD to automatically correct labels. We then apply BCD to event detection to identify transitions between appliances’ on and off states. Performance is evaluated using 3 publicly available datasets containing over 250 appliances across 11 houses. Results show both BCD applications are competitive and in some cases outperform existing state-of-the-art methods without the need for parameter tuning, advancing disaggregation towards widespread, real-world deployment. CCS Concepts •Theory of computation → Theory and algorithms for application domains; •Social and professional topics → Sustainability;

AAAI Conference 2017 Conference Paper

Multi-Robot Allocation of Tasks with Temporal and Ordering Constraints

• Maria Gini

Task allocation is ubiquitous in computer science and robotics, yet some problems have received limited attention in the computer science and AI community. Specifically, we will focus on multi-robot task allocation problems when tasks have time windows or ordering constraints. We will outline the main lines of research and open problems.

AAMAS Conference 2017 Conference Paper

Safely Using Predictions in General-Sum Normal Form Games

• Steven Damer
• Maria Gini

It is often useful to predict opponent behavior when playing a generalsum two-player normal form game. However best-responding to an inaccurate prediction can lead to a strategy which is vulnerable to exploitation. This paper proposes a novel method, Restricted Stackelberg Response with Safety (RSRS), for an agent to select a strategy to respond to a prediction. The agent uses the confidence it has in the prediction and a safety margin which reflects the level of risk it is willing to tolerate to make a controlled trade-off between best-responding to the prediction and providing a guarantee of worst-case performance. We describe an algorithm which selects parameter values for RSRS to produce strategies that play well against the prediction, respond to a best-responding opponent, and guard against worst-case outcomes. We report results obtained by the algorithm on multiple general-sum games against different opponents.

IJCAI Conference 2016 Conference Paper

Controlling Growing Tasks with Heterogeneous Agents

• James Parker
• Maria Gini

We propose solutions for assignment of physical tasks to heterogeneous agents when the costs of the tasks change over time. We assume tasks have a natural growth rate which is counteracted by the work applied by the agents. As the future cost of a task depends on the agents allocation, reasoning must be both spatial and temporal to effectively minimize the growth so tasks can be completed. We present optimal solutions for two general classes of growth functions and heuristic solutions for other cases. Empirical results are given in RoboCup Rescue for agents with different capabilities.

AAAI Conference 2016 Conference Paper

Implicit Coordination in Crowded Multi-Agent Navigation

• Julio Godoy
• Ioannis Karamouzas
• Stephen Guy
• Maria Gini

In crowded multi-agent navigation environments, the motion of the agents is significantly constrained by the motion of the nearby agents. This makes planning paths very difficult and leads to inefficient global motion. To address this problem, we propose a new distributed approach to coordinate the motions of agents in crowded environments. With our approach, agents take into account the velocities and goals of their neighbors and optimize their motion accordingly and in real-time. We experimentally validate our coordination approach in a variety of scenarios and show that its performance scales to scenarios with hundreds of agents.

AAMAS Conference 2016 Conference Paper

Iterated Multi-Robot Auctions for Precedence-Constrained Task Scheduling

• Mitchell McIntire
• Ernesto Nunes
• Maria Gini

AAAI Conference 2016 Conference Paper

Monte Carlo Tree Search for Multi-Robot Task Allocation

• Bilal Kartal
• Ernesto Nunes
• Julio Godoy
• Maria Gini

Multi-robot teams are useful in a variety of task allocation domains such as warehouse automation and surveillance. Robots in such domains perform tasks at given locations and specific times, and are allocated tasks to optimize given team objectives. We propose an efficient, satisficing and centralized Monte Carlo Tree Search based algorithm exploiting branch and bound paradigm to solve the multi-robot task allocation problem with spatial, temporal and other side constraints. Unlike previous heuristics proposed for this problem, our approach offers theoretical guarantees and finds optimal solutions for some non-trivial data sets.

IJCAI Conference 2016 Conference Paper

Moving in a Crowd: Safe and Efficient Navigation among Heterogeneous Agents

• Julio Godoy
• Ioannis Karamouzas
• Stephen J. Guy
• Maria Gini

Multi-agent navigation methods typically assume that all agents use the same underlying framework to navigate to their goal while avoiding colliding with each other. However, such assumption does not hold when agents do not know how other agents will move. We address this issue by proposing a Bayesian inference approach where an agent estimates the navigation model and goal of each neighbor, and uses this to compute a plan that minimizes collisions while driving it to its goal. Simulation experiments performed in many scenarios demonstrate that an agent using our approach computes safer and more time-efficient paths as compared to those generated without our inference approach anda state-of-the-art local navigation framework.

IJCAI Conference 2015 Conference Paper

Mining Expert Play to Guide Monte Carlo Search in the Opening Moves of Go

• Erik S. Steinmetz
• Maria Gini

We propose a method to guide a Monte Carlo search in the initial moves of the game of Go. Our method matches the current state of a Go board against clusters of board configurations that are derived from a large number of games played by experts. The main advantage of this method is that it does not require an exact match of the current board, and hence is effective for a longer sequence of moves compared to traditional opening books. We apply this method to two different open-source Go-playing programs. Our experiments show that this method, through its filtering or biasing the choice of a next move to a small subset of possible moves, improves play effectively in the initial moves of a game.

AAAI Conference 2015 Conference Paper

Multi-Robot Auctions for Allocation of Tasks with Temporal Constraints

• Ernesto Nunes
• Maria Gini

We propose an auction algorithm to allocate tasks that have temporal constraints to cooperative robots. Temporal constraints are expressed as time windows, within which a task must be executed. There are no restrictions on the time windows, which are allowed to overlap. Robots model their temporal constraints using a simple temporal network, enabling them to maintain consistent schedules. When bidding on a task, a robot takes into account its own current commitments and an optimization objective, which is to minimize the time of completion of the last task alone or in combination with minimizing the distance traveled. The algorithm works both when all the tasks are known upfront and when tasks arrive dynamically. We show the performance of the algorithm in simulation with different numbers of tasks and robots, and compare it with a baseline greedy algorithm and a state-ofthe-art auction algorithm. Our algorithm is computationally frugal and consistently allocates more tasks than the competing algorithms.

TIST Journal 2015 Journal Article

On Optimizing Airline Ticket Purchase Timing

• William Groves
• Maria Gini

Proper timing of the purchase of airline tickets is difficult even when historical ticket prices and some domain knowledge are available. To address this problem, we introduce an algorithm that optimizes purchase timing on behalf of customers and provides performance estimates of its computed action policy. Given a desired flight route and travel date, the algorithm uses machine-learning methods on recent ticket price quotes from many competing airlines to predict the future expected minimum price of all available flights. The main novelty of our algorithm lies in using a systematic feature-selection technique, which captures time dependencies in the data by using time-delayed features, and reduces the number of features by imposing a class hierarchy among the raw features and pruning the features based on in-situ performance. Our algorithm achieves much closer to the optimal purchase policy than other existing decision theoretic approaches for this domain, and meets or exceeds the performance of existing feature-selection methods from the literature. Applications of our feature-selection process to other domains are also discussed.

AAMAS Conference 2013 Conference Paper

An Agent for Optimizing Airline Ticket Purchasing

• William Groves
• Maria Gini

Buying airline tickets is an ubiquitous task in which it is difficult for humans to minimize cost due to insufficient information. Even with historical data available for inspection (a recent addition to some travel reservation websites), it is difficult to assess how purchase timing translates into changes in expected cost. To address this problem, we introduce an agent which is able to optimize purchase timing on behalf of customers. We provide results that demonstrate the method can perform much closer to the optimal purchase policy than existing decision theoretic approaches for this domain.

IJCAI Conference 2013 Conference Paper

Optimal Airline Ticket Purchasing Using Automated User-Guided Feature Selection

• William Groves
• Maria Gini

Airline ticket purchase timing is a strategic problem that requires both historical data and domain knowledge to solve consistently. Even with some historical information (often a feature of modern travel reservation web sites), it is difficult for consumers to make true cost-minimizing decisions. To address this problem, we introduce an automated agent which is able to optimize purchase timing on behalf of customers and provide performance estimates of its computed action policy based on past performance. We apply machine learning to recent ticket price quotes from many competing airlines for the target flight route. Our novelty lies in extending this using a systematic feature extraction technique incorporating elementary user-provided domain knowledge that greatly enhances the performance of machine learning algorithms. Using this technique, our agent achieves much closer to the optimal purchase policy than other proposed decision theoretic approaches for this domain.

IJCAI Conference 2013 Conference Paper

Rolling Dispersion for Robot Teams

• Elizabeth A. Jensen
• Maria Gini

Dispersing a team of robots into an unknown and dangerous environment, such as a collapsed building, can provide information about structural damage and locations of survivors and help rescuers plan their actions. We propose a rolling dispersion algorithm, which makes use of a small number of robots and achieves full exploration. The robots disperse as much as possible while maintaining communication, and then advance as a group, leaving behind beacons to mark explored areas and provide a path back to the entrance. The novelty of this algorithm comes from the manner in which the robots continue their exploration as a group after reaching the maximum dispersion possible while staying in contact with each other. We use simulation to show that the algorithm works in multiple environments and for varying numbers of robots.

AAMAS Conference 2012 Conference Paper

Auctioning Robotic Tasks with Overlapping Time Windows

• Ernesto Nunes
• Maitreyi Nanjanath
• Maria Gini

This work investigates allocation of tasks to multi-robots when tasks are spatially distributed and constrained to be executed within assigned time windows. Our work explores the interaction between scheduling and optimal routing. We propose the Time-Sensitive Sequential Single-Item Auction algorithm as a method to allocate tasks with time windows in multi-robot systems. We show, experimentally, that the proposed algorithm outperforms other auction algorithms that we modified to handle time windows.

AAMAS Conference 2012 Conference Paper

When speed matters in learning against adversarial opponents

• Mohamed Elidrisi
• Maria Gini

We propose a novel algorithm that is able to learn and adapt to an opponent even within a limited number of interactions and against a rapidly adapting opponent. The context we use is two player normal form games. We compare the performance of an agent using our algorithm against agents using existing multiagent learning algorithms.

AAMAS Conference 2011 Conference Paper

Friend or Foe? Detecting an Opponent's Attitude in Normal Form Games

• Steven Damer
• Maria Gini

We study the problem of achieving cooperation between two self-interested agents that play a sequence of different randomly generated normal form games. The agent learns how much the opponent is willing to cooperate and reciprocates. We present empirical results that show that both agents benefit from cooperation and that a small number of games is sufficient to learn the cooperation level of the opponent.

AAMAS Conference 2010 Conference Paper

MAITH: a Meta-software Agent for Issue Tracking Help

• Touby Drew
• Maria Gini

Issue tracking is an essential part of regulated software development where it is typically supported by software systemswhich are complex and not easily customizable. We proposea meta-software agent that senses what windows and widgets are in focus by the user and leverages this awarenessto provide support. The user is given ways of making andrecalling annotations appropriate for the context. By observing users in action the agent creates models which canthen be used to predict and suggest next steps. This paper describes an early prototype of this approach built asa proof of concept. Preliminary results and directions forfuture work are outlined.

AAMAS Conference 2006 Conference Paper

Implantable Medical Devices as Agents and Part of Multiagent Systems

• Touby Drew
• Maria Gini

AAAI Conference 2005 Conference Paper

Non-Stationary Policy Learning in 2-Player Zero Sum Games

• Steven Jensen
• Maria Gini

A key challenge in multiagent environments is the construction of agents that are able to learn while acting in the presence of other agents that are simultaneously learning and adapting. These domains require on-line learning methods without the benefit of repeated training examples, as well as the ability to adapt to the evolving behavior of other agents in the environment. The difficulty is further exacerbated when the agents are in an adversarial relationship, demanding that a robust (i. e. winning) non-stationary policy be rapidly learned and adapted. We propose an on-line sequence learning algorithm, ELPH, based on a straightforward entropy pruning technique that is able to rapidly learn and adapt to non-stationary policies. We demonstrate the performance of this method in a non-stationary learning environment of adversarial zero-sum matrix games.

AAMAS Conference 2005 Conference Paper

Rapid On-line Temporal Sequence Prediction by an Adaptive Agent

• Steven Jensen
• Daniel Boley
• Maria Gini
• Paul Schrater

AAMAS Conference 2004 Conference Paper

Harnessing the Search for Bid Schedules with Stochastic Search and Domain-specific Heuristics

• Alexander Babanov
• John Collins
• Maria Gini

AAMAS Conference 2003 Conference Paper

An Evolutionary Framework for Studying Behaviors of Economic Agents

• Wolfgang Ketter
• Alexander Babanov
• Maria Gini

AAMAS Conference 2003 Conference Paper

Scheduling Tasks with Precedence Constraints to Solicit Desirable Bid Combinations

• Alexander Babanov
• John Collins
• Maria Gini

AIIM Journal 1997 Journal Article

Diagnosing congenital heart defects using the Fallot computational model

• Nancy E. Reed
• Maria Gini
• Paul E. Johnson
• James H. Moller

This paper describes a computational model developed for the diagnosis of multiple defects. If multiple defects interact, meaning that the cues observable for multiple defects are not a sum of the cues observable for the component defects, diagnosis is particularly difficult. We developed a description and classification of the ways cues change when defects interact. A computational model (named Fallot) was implemented and a knowledge-base was constructed for the diagnosis of congenital heart defects. On each case, Fallot performs recognition-based reasoning followed by solution construction and evaluation with the cue combination methods. Fallot was tested on cases from hospital files and correctly diagnoses cases with multiple interacting defects for which conventional methods are not applicable or fail.

IJCAI Conference 1983 Conference Paper

Towards Automatic Error Recovery in Robot Programs

• Maria Gini
• Giuseppina Gini