Héctor Muñoz-Avila

Lehigh University, Bethlehem, Pennsylvania, United States

Are you Héctor Muñoz-Avila?

Claim your profile

Publications (98)15.34 Total impact

  • Source
    Pinar Öztürk, Hector Munoz-Avila, Agnar Aamodt
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper discusses the need to recognize and take advantage of arising opportunities when an agent is autonomously interacting with an environment. We discuss the tasks, methods, model framework, and use it to analyze opportunities across several dimensions: when changing the tasks versus changing the methods, when having a complete knowledge about the methods in the domain versus having partial knowledge, when having a complete model of the domain versus having an incomplete model, when tackling a single task versus multiple tasks and finally we discuss the role of causal explanations in opportunistic case-based decision making.
    22nd International Conference on Case-based Reasoning, Cork, Ireland; 09/2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Cervical cancer is the second most common type of cancer for women. Existing screening programs for cervical cancer, such as Pap Smear, suffer from low sensitivity. Thus, many patients who are ill are not detected in the screening process. Using images of the cervix as an aid in cervical cancer screening has the potential to greatly improve sensitivity, and can be especially useful in resource-poor regions of the world. In this work, we develop a data-driven computer algorithm for interpreting cervical images based on color and texture. We are able to obtain 74% sensitivity and 90% specificity when differentiating high-grade cervical lesions from low-grade lesions and normal tissue. On the same dataset, using Pap tests alone yields a sensitivity of 37% and specificity of 96%, and using HPV test alone gives a 57% sensitivity and 93% specificity. Furthermore, we develop a comprehensive algorithmic framework based on Multi-Modal Entity Coreference for combining various tests to perform disease classification and diagnosis. When integrating multiple tests, we adopt information gain and gradient-based approaches for learning the relative weights of different tests. In our evaluation, we present a novel algorithm that integrates cervical images, Pap, HPV and patient age, which yields 83.21% sensitivity and 94.79% specificity, a statistically significant improvement over using any single source of information alone.
    IEEE transactions on medical imaging. 08/2014;
  • Hankz Hankui Zhuo, Héctor Muñoz-Avila, Qiang Yang
    [Show abstract] [Hide abstract]
    ABSTRACT: Hierarchical Task Network (HTN) planning is an effective yet knowledge intensive problem-solving technique. It requires humans to encode knowledge in the form of methods and action models. Methods describe how to decompose tasks into subtasks and the preconditions under which those methods are applicable whereas action models describe how actions change the world. Encoding such knowledge is a difficult and time-consuming process, even for domain experts. In this paper, we propose a new learning algorithm, called HTNLearn, to help acquire HTN methods and action models. HTNLearn receives as input a collection of plan traces with partially annotated intermediate state information, and a set of annotated tasks that specify the conditions before and after the tasks' completion. In addition, plan traces are annotated with potentially empty partial decomposition trees that record the processes of decomposing tasks to subtasks. HTNLearn outputs are a collection of methods and action models. HTNLearn first encodes constraints about the methods and action models as a constraint satisfaction problem, and then solves the problem using a weighted MAX-SAT solver. HTNLearn can learn methods and action models simultaneously from partially observed plan traces (i.e., plan traces where the intermediate states are partially observable). We test HTNLearn in several HTN domains. The experimental results show that our algorithm HTNLearn is both effective and efficient.
    Artificial Intelligence 01/2014; · 2.19 Impact Factor
  • Source
    Dustin Dannenhauer, Héctor Muñoz-Avila
    [Show abstract] [Hide abstract]
    ABSTRACT: IBM's Watson uses a variety of scoring algorithms to rank candidate answers for natural language questions. These scoring algorithms played a crucial role in Watson's win against human champions in Jeopardy!. We show that this same technique can be implemented within a real-time strategy (RTS) game playing goal-driven autonomy (GDA) agent. Previous GDA agents in RTS games were forced to use very compact state representations. Watson's scoring algorithms tech-nique removes this restriction for goal selection, allowing the use of all information available in the game state. Unfortunately, there is a high knowledge engineering effort required to create new scoring algorithms. We alleviate this burden using case-based reasoning to approximate past observations of a scoring algorithm system. Our experiments in a real-time strategy game show that goal selection by the CBR system attains comparable in-game performance to a baseline scoring algorithm system.
    International Conference On Case-based Reasoning; 07/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: The incorporation of learning into commercial games can enrich the player experience, but may concern developers in terms of issues such as losing control of their game world. We explore a number of applied research and some fielded applications that point to the tremendous possibilities of machine learning research including game genres such as real-time strategy games, flight simulation games, car and motorcycle racing games, board games such as Go, an even traditional game-theoretic problems such as the prisoners dilemma. A common trait of these works is the potential of machine learning to reduce the burden of game developers. However a number of challenges exists that hinder the use of machine learning more broadly. We discuss some of these challenges while at the same time exploring opportunities for a wide use of machine learning in games.
    Artificial and Computational Intelligence in Games, Edited by Simon M. Lucas, Michael Mateas, Mike Preuss, Pieter Spronck, Julian Togelius, 01/2013: chapter Learning and Game AI: pages 33-43; Dagstuhl Publishing., ISBN: 978-3-939897-62-0
  • Source
    Ulit Jaidee, Héctor Muñoz-Avila, David W Aha
    [Show abstract] [Hide abstract]
    ABSTRACT: We describe a study on using case-based learning techniques in a goal-driven autonomy (GDA) agent for real-time strategy games. The two case bases in our Learning GDA (LGDA) agent store (1) the expected states that an agent can reach when executing an action in and (2) the next goals to pursue when a discrepancy occurs between the expected and encountered states. We report on an ablation study that demonstrates performance gains using LGDA.
    01/2011;
  • Source
    Hankz Hankui Zhuo, Hector Muñoz-Avila, Qiang Yang
    [Show abstract] [Hide abstract]
    ABSTRACT: In multi-agent planning environments, action models for each agent must be given as input. However, creating such action models by hand is difficult and time-consuming, because it requires formally representing the complex relationships among different objects in the environment. The problem is compounded in multi-agent environments where agents can take more types of actions. In this paper, we present an algorithm to learn action models for multi-agent planning systems from a set of input plan traces. Our learning algorithm Lammas automatically generates three kinds of constraints: (1) constraints on the interactions between agents, (2) constraints on the correctness of the action models for each individual agent, and (3) constraints on actions themselves. Lammas attempts to satisfy these constraints simultaneously using a weighted maximum satisfiability model known as MAX-SAT, and converts the solution into action models. We believe this to be one of the first learning algorithms to learn action models in the context of multi-agent planning environments. We empirically demonstrate that Lammas performs effectively and efficiently in several planning domains.
    10th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2011), Taipei, Taiwan, May 2-6, 2011, Volume 1-3; 01/2011
  • Source
    Alexandra Coman, Hector Muñoz-Avila
    Proceedings of the Twenty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2011, San Francisco, California, USA, August 7-11, 2011; 01/2011
  • Ulit Jaidee, Héctor Muñoz-Avila, David W. Aha
    [Show abstract] [Hide abstract]
    ABSTRACT: Goal-driven autonomy (GDA) is a reflective model of goal reasoning that controls the focus of an agent's planning activities by dynamically resolving unexpected discrepancies in the world state, which frequently arise when solving tasks in complex environments. GDA agents have performed well on such tasks by integrating methods for discrepancy recognition, explanation, goal formulation, and goal management. However, they require substantial domain knowledge, including what constitutes a discrepancy and how to resolve it. We introduce LGDA, a learning algorithm for acquiring this knowledge, modeled as cases, that and integrates case-based reasoning and reinforcement learning methods. We assess its utility on tasks from a complex video game environment. We claim that, for these tasks, LGDA can significantly outperform its ablations. Our evaluation provides evidence to support this claim. LGDA exemplifies a feasible design methodology for deployable GDA agents.
    IJCAI 2011, Proceedings of the 22nd International Joint Conference on Artificial Intelligence, Barcelona, Catalonia, Spain, July 16-22, 2011; 01/2011
  • Source
    Alexandra Coman, Héctor Muñoz-Avila
    [Show abstract] [Hide abstract]
    ABSTRACT: The concept of diversity was successfully introduced for recommender-systems. By displaying results that are not only similar to a target problem but also diverse among themselves, recommender systems have been shown to provide more effective guidance to the user. We believe that similar benefits can be obtained in case-based planning, provided that diversity-enhancement techniques can be adapted appropriately. Our claim is that diversity is truly useful when it refers not only to the initial and goal states of a plan, but also to the sequence of actions the plan consists of. To formalize this characteristic and support our claim, we define the metric of “plan diversity” and put it to test using plans for a real-time strategy game, a domain chosen for the simplicity and clarity of its tasks and the quantifiable results it generates.
    Case-Based Reasoning Research and Development - 19th International Conference on Case-Based Reasoning, ICCBR 2011, London, UK, September 12-15, 2011. Proceedings; 01/2011
  • Source
    Héctor Muñoz-Avila, Ulit Jaidee, David W. Aha, Elizabeth Carter
    [Show abstract] [Hide abstract]
    ABSTRACT: The vast majority of research on AI planning has focused on automated plan recognition, in which a planning agent is provided with a set of inputs that include an initial goal (or set of goals). In this context, the goal is presumed to be static; it never changes, and the agent is not provided with the ability to reason about whether it should change this goal. For some tasks in complex environments, this constraint is problematic; the agent will not be able to respond to opportunities or plan execution failures that would benefit from focusing on a different goal. Goal driven autonomy (GDA) is a reasoning framework that was recently introduced to address this limitation; GDA systems perform anytime reasoning about what goal(s) should be satisfied [4]. Although promising, there are natural roles that case-based reasoning (CBR) can serve in this framework, but no such demonstration exists. In this paper, we describe the GDA framework and describe an algorithm that uses CBR to support it. We also describe an empirical study with a multiagent gaming environment in which this CBR algorithm outperformed a rule-based variant of GDA as well as a non-GDA agent that is limited to dynamic replanning.
    Case-Based Reasoning. Research and Development, 18th International Conference on Case-Based Reasoning, ICCBR 2010, Alessandria, Italy, July 19-22, 2010. Proceedings; 01/2010
  • Source
    Proceedings of the Twenty-Third International Florida Artificial Intelligence Research Society Conference, May 19-21, 2010, Daytona Beach, Florida; 01/2010
  • Source
    Matt Dilts, Héctor Muñoz-Avila
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we present an approach for reducing the memory footprint requirement of temporal difference methods in which the set of states is finite. We use case-based generalization to group the states visited during the reinforcement learning process. We follow a lazy learning approach; cases are grouped in the order in which they are visited. Any new state visited is assigned to an existing entry in the Q-table provided that a similar state has been visited before. Otherwise a new entry is added to the Q-table. We performed experiments on a turn-based game where actions have non-deterministic effects and might have long term repercussions on the outcome of the game. The main conclusion from our experiments is that by using case-based generalization, the size of the Q-table can be substantially reduced while maintaining the quality of the RL estimates.
    Case-Based Reasoning. Research and Development, 18th International Conference on Case-Based Reasoning, ICCBR 2010, Alessandria, Italy, July 19-22, 2010. Proceedings; 01/2010
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we study the topic of CBR systems learning from observations in which those observations can be represented as stochastic policies. We describe a general framework which encompasses three steps: (1) it observes agents performing actions, elicits stochastic policies representing the agents’ strategies and retains these policies as cases. (2) The agent analyzes the environment and retrieves a suitable stochastic policy. (3) The agent then executes the retrieved stochastic policy, which results in the agent mimicking the previously observed agent. We implement our framework in a system called JuKeCB that observes and mimics players playing games. We present the results of three sets of experiments designed to evaluate our framework. The first experiment demonstrates that JuKeCB performs well when trained against a variety of fixed strategy opponents. The second experiment demonstrates that JuKeCB can also, after training, win against an opponent with a dynamic strategy. The final experiment demonstrates that JuKeCB can win against "new" opponents (i.e. opponents against which JuKeCB is untrained).
    Case-Based Reasoning. Research and Development, 18th International Conference on Case-Based Reasoning, ICCBR 2010, Alessandria, Italy, July 19-22, 2010. Proceedings; 01/2010
  • Source
    Ugur Kuter, Hector Muñoz-Avila
    AI Magazine. 01/2010; 31:101-102.
  • Source
    Chad Hogg, Ugur Kuter, Hector Muñoz-Avila
    [Show abstract] [Hide abstract]
    ABSTRACT: We consider how to learn Hierarchical Task Networks (HTNs) for planning problems in which both the quality of solution plans generated by the HTNs and the speed at which those plans are found is important. We describe an integration of HTN Learning with Reinforcement Learning to both learn methods by analyzing semantic annotations on tasks and to produce estimates of the expected values of the learned meth- ods by performing Monte Carlo updates. We performed an experiment in which plan quality was inversely related to plan length. In two planning domains, we evaluated the planning performance of the learned methods in comparison to two state-of-the-art satisficing classical planners, FASTFORWARD and SGPLAN6, and one optimal planner, HSP * F. The results demonstrate that a greedy HTN planner using the learned methods was able to generate higher quality solutions than SGPLAN6 in both domains and FASTFORWARD in one. Our planner, FASTFORWARD, and SGPLAN6 ran in similar time, while HSP * F was exponentially slower.
    Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence, AAAI 2010, Atlanta, Georgia, USA, July 11-15, 2010; 01/2010
  • Source
    Stephen Lee-Urban, Héctor Muñoz-Avila
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we revisit the trade-off between adaptation and retrieval effort traditionally held as a principle in case-based reasoning. This principle states that the time needed for adaptation reduces with the time spent searching for an adequate case to be retrieved. In particular, if very little time is spent in retrieval, the adaptation effort will be high. Correspondingly, if the retrieval effort is high, the adaption effort is low. We analyzed this principle in two boundary conditions: (1) when very bad and (2) when highly capable adaptation procedures are used. We conclude that in the first boundary condition the adaptation-retrieval trade-off does not necessarily exist. We also claim that the second does not hold for a class of planning domains frequently used in the literature. To validate this claim, we performed experiments on two domains of this type.
    Case-Based Reasoning Research and Development, 8th International Conference on Case-Based Reasoning, ICCBR 2009, Seattle, WA, USA, July 20-23, 2009, Proceedings; 01/2009
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To apply hierarchical task network (HTN) plan- ning to real-world planning problems, one needs to encode the HTN schemata and action mod- els beforehand. However, acquiring such domain knowledge is difficult and time-consuming because the HTN domain definition involves a significant knowledge-engineering effort. A system that can learn the HTN planning domain knowledge auto- matically would save time and allow HTN plan- ning to be used in domains where such knowledge- engineering effort is not feasible. In this paper, we present a formal framework and algorithms to ac- quire HTN planning domain knowledge, by learn- ing the preconditions and effects of actions and preconditions of methods. Our algorithm, HTN- learner, first builds constraints from given observed decomposition trees to build action models and method preconditions. It then solves these con- straints using a weighted MAX-SAT solver. The solution can be converted to action models and method preconditions. Unlike prior work on HTN learning, we do not depend on complete action models or state information. We test the algorithm on several domains, and show that our HTN-learner algorithm is both effective and efficient.
    IJCAI 2009, Proceedings of the 21st International Joint Conference on Artificial Intelligence, Pasadena, California, USA, July 11-17, 2009; 01/2009
  • Chad Hogg, Ugur Kuter, Héctor Muñoz-Avila
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper describes how to learn Hierarchical Task Networks (HTNs) in nondeterministic planning do- mains, where actions may have multiple possi- ble outcomes. We discuss several desired prop- erties that guarantee that the resulting HTNs will correctly handle the nondeterminism in the do- main. We developed a new learning algorithm, called HTN-MAKERND , that exploits these prop- erties. We implemented HTN-MAKERND in the recently-proposed HTN-MAKER system, a goal- regression based HTN learning approach. In our theoretical study, we show that HTN-MAKERND soundly produces HTN planning knowledge in low-order polynomial times, despite the nondeter- minism. In our experiments with two nondetermin- istic planning domains, ND-SHOP2, a well-known HTN planning algorithm for nondeterministic do- mains, significantly outperformed (in some cases, by about 3 orders of magnitude) the well-known planner MBP using the learned HTNs.
    IJCAI 2009, Proceedings of the 21st International Joint Conference on Artificial Intelligence, Pasadena, California, USA, July 11-17, 2009; 01/2009
  • Source
    Hua Li, Héctor Muñoz-Avila, Diane Bramsen, Chad Hogg, Rafael Alonso
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a new approach for spatial event prediction that combines a value function approximation algorithm and case-based reasoning predictors. Each of these predictors makes unique contributions to the overall spatial event prediction. The function value approximation prediction is particularly suitable to reasoning with geographical features such as the (x,y) coordinates of an event. The case-based prediction is particularly well suited to deal with non-geographical features such as the time of the event or income level of the population. We claim that the combination of these two predictors results in a significant improvement of the accuracy in the spatial event prediction compared to pure geographically-based predictions. We support our claim by reporting on an ablation study for the prediction of improvised explosive device (IED) attacks.
    Case-Based Reasoning Research and Development, 8th International Conference on Case-Based Reasoning, ICCBR 2009, Seattle, WA, USA, July 20-23, 2009, Proceedings; 01/2009

Publication Stats

1k Citations
15.34 Total Impact Points

Institutions

  • 2002–2014
    • Lehigh University
      • Department of Computer Science and Engineering
      Bethlehem, Pennsylvania, United States
  • 1999–2003
    • University of Maryland, College Park
      • Department of Computer Science
      Maryland, United States
  • 2001
    • Loyola University Maryland
      Baltimore, Maryland, United States
    • Fachhochschule Aachen
      Aachen, North Rhine-Westphalia, Germany