dung tran - Academia.edu (original) (raw)

Papers by dung tran

Research paper thumbnail of Designing massive open online courses for educators: The case of teaching statistics

Research paper thumbnail of The Role of Probability in Developing Learners’ Models of Simulation Approaches to Inference

STATISTICS EDUCATION RESEARCH JOURNAL, 2016

Repeated sampling approaches to inference that rely on simulations have recently gained prominenc... more Repeated sampling approaches to inference that rely on simulations have recently gained prominence in statistics education, and probabilistic concepts are at the core of this approach. In this approach, learners need to develop a mapping among the problem situation, a physical enactment, computer representations, and the underlying randomization and sampling processes. We explicate the role of probability in this approach and draw upon a models and modeling perspective to support the development of teachers’ models for using a repeated sampling approach for inference. We explicate the model development task sequence and examine the teachers’ representations of their conceptualizations of a repeated sampling approach for inference. We propose key conceptualizations that can guide instruction when using simulations and repeated sampling for drawing inferences. First published November 2016 at Statistics Education Research Journal Archives

Research paper thumbnail of Factored State-Abstract Hidden Markov Models for Activity Recognition Using Pervasive Multi-modal Sensors

2005 International Conference on Intelligent Sensors, Sensor Networks and Information Processing, 2005

Current probabilistic models for activity recognition do not incorporate much sensory input data ... more Current probabilistic models for activity recognition do not incorporate much sensory input data due to the problem of state space explosion. In this paper, we propose a model for activity recognition, called the Factored State-Abtract Hidden Markov Model (FS-AHMM) to allow us to integrate many sensors for improving recognition performance. The proposed FS-AHMM is an extension of the Abstract Hidden Markov Model which applies the concept offactored state representations to compactly represent the state transitions. The parameters of the FS-AHMM are estimated using the EM algorithm from the data acquired through multiple multi-modal sensors and cameras. The model is evaluated and compared with other exisiting models on real-world data. The results show that the proposed model outperforms other models and that the integrated sensor information helps in recognizing activity more accurately.

Research paper thumbnail of A probabilistic model with parsinomious representation for sensor fusion in recognizing activity in pervasive environment

18th International Conference on Pattern Recognition (ICPR'06), 2006

To tackle the problem of increasing numbers of state transition parameters when the number of sen... more To tackle the problem of increasing numbers of state transition parameters when the number of sensors increases, we present a probabilistic model together with several parsinomious representations for sensor fusion. These include context specific independence (CSI), mixtures of smaller multinomials and softmax function representations to compactly represent the state transitions of a large number of sensors. The model is evaluated on real-world data acquired through ubiquitous sensors in recognizing daily morning activities. The results show that the combination of CSI and mixtures of smaller multinomials achieves comparable performance with much fewer parameters.

Research paper thumbnail of Tank War Using Online Reinforcement Learning

2009 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology, 2009

Real-Time Strategy(RTS) games provide a challenging platform to implement online reinforcement le... more Real-Time Strategy(RTS) games provide a challenging platform to implement online reinforcement learning(RL) techniques in a real application. Computer as one player monitors opponents'(human or other computers) strategies and then updates its own policy using RL methods. In this paper, we propose a multi-layer framework for implementing the online RL in a RTS game. The framework significantly reduces the RL computational complexity by decomposing the state space in a hierarchical manner. We implement the RTS game-Tank General, and perform a thorough test on the proposed framework. The results show the effectiveness of our proposed framework and shed light on relevant issues on using the RL in RTS games.

Research paper thumbnail of An argument-based approach to reasoning with specificity

Artificial Intelligence, 2001

We present a new priority-based approach to reasoning with specificity which subsumes inheritance... more We present a new priority-based approach to reasoning with specificity which subsumes inheritance reasoning. The new approach differs from other priority-based approaches in the literature in the way priority between defaults is handled. Here, it is conditional rather than unconditional as in other approaches. We show that any unconditional handling of priorities between defaults as advocated in the literature until now is not sufficient to capture general defeasible inheritance reasoning. We propose a simple and novel argumentation semantics for reasoning with specificity taking the conditionality of the priorities between defaults into account. Since the proposed argumentation semantics is a form of stable semantics of nonmonotonic reasoning, it inherits a common problem of the latter where it is not always defined for every default theory. We propose a class of stratified default theories for which the argumentation semantics is always defined. We also show that acyclic and consistent inheritance networks are stratified. We prove that the argumentation semantics satisfies the basic properties of a nonmonotonic consequence relation such as deduction, reduction, conditioning, and cumulativity for well-defined and stratified default theories. We give a modular and polynomial transformation of default theories with specificity into semantically equivalent Reiter default theories.

Research paper thumbnail of Experiments with Online Reinforcement Learning in Real-Time Strategy Games

Applied Artificial Intelligence, 2009

Real-Time Strategy (RTS) games provide a challenging platform to implement online reinforcement l... more Real-Time Strategy (RTS) games provide a challenging platform to implement online reinforcement learning (RL) techniques in a real application. Computer as one player monitors opponents' (human or other computers) strategies and then updates its own policy using RL methods. In this paper, we firstly examine the suitability of applying the online RL in various computer games. RL application depends much on both RL complexity and the game features. We then propose a multi-layer framework for implementing online RL in an RTS game. The framework significantly reduces RL computational complexity by decomposing the state space in a hierarchical manner. We implement an RTS game-Tank General, and perform a thorough test on the proposed framework. We consider three typical profiles of RTS game players and compare two basic RL techniques applied in the game. The results show the effectiveness of our proposed framework and shed light on relevant issues on using online RL in RTS games.

Research paper thumbnail of Designing massive open online courses for educators: The case of teaching statistics

Research paper thumbnail of The Role of Probability in Developing Learners’ Models of Simulation Approaches to Inference

STATISTICS EDUCATION RESEARCH JOURNAL, 2016

Repeated sampling approaches to inference that rely on simulations have recently gained prominenc... more Repeated sampling approaches to inference that rely on simulations have recently gained prominence in statistics education, and probabilistic concepts are at the core of this approach. In this approach, learners need to develop a mapping among the problem situation, a physical enactment, computer representations, and the underlying randomization and sampling processes. We explicate the role of probability in this approach and draw upon a models and modeling perspective to support the development of teachers’ models for using a repeated sampling approach for inference. We explicate the model development task sequence and examine the teachers’ representations of their conceptualizations of a repeated sampling approach for inference. We propose key conceptualizations that can guide instruction when using simulations and repeated sampling for drawing inferences. First published November 2016 at Statistics Education Research Journal Archives

Research paper thumbnail of Factored State-Abstract Hidden Markov Models for Activity Recognition Using Pervasive Multi-modal Sensors

2005 International Conference on Intelligent Sensors, Sensor Networks and Information Processing, 2005

Current probabilistic models for activity recognition do not incorporate much sensory input data ... more Current probabilistic models for activity recognition do not incorporate much sensory input data due to the problem of state space explosion. In this paper, we propose a model for activity recognition, called the Factored State-Abtract Hidden Markov Model (FS-AHMM) to allow us to integrate many sensors for improving recognition performance. The proposed FS-AHMM is an extension of the Abstract Hidden Markov Model which applies the concept offactored state representations to compactly represent the state transitions. The parameters of the FS-AHMM are estimated using the EM algorithm from the data acquired through multiple multi-modal sensors and cameras. The model is evaluated and compared with other exisiting models on real-world data. The results show that the proposed model outperforms other models and that the integrated sensor information helps in recognizing activity more accurately.

Research paper thumbnail of A probabilistic model with parsinomious representation for sensor fusion in recognizing activity in pervasive environment

18th International Conference on Pattern Recognition (ICPR'06), 2006

To tackle the problem of increasing numbers of state transition parameters when the number of sen... more To tackle the problem of increasing numbers of state transition parameters when the number of sensors increases, we present a probabilistic model together with several parsinomious representations for sensor fusion. These include context specific independence (CSI), mixtures of smaller multinomials and softmax function representations to compactly represent the state transitions of a large number of sensors. The model is evaluated on real-world data acquired through ubiquitous sensors in recognizing daily morning activities. The results show that the combination of CSI and mixtures of smaller multinomials achieves comparable performance with much fewer parameters.

Research paper thumbnail of Tank War Using Online Reinforcement Learning

2009 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology, 2009

Real-Time Strategy(RTS) games provide a challenging platform to implement online reinforcement le... more Real-Time Strategy(RTS) games provide a challenging platform to implement online reinforcement learning(RL) techniques in a real application. Computer as one player monitors opponents'(human or other computers) strategies and then updates its own policy using RL methods. In this paper, we propose a multi-layer framework for implementing the online RL in a RTS game. The framework significantly reduces the RL computational complexity by decomposing the state space in a hierarchical manner. We implement the RTS game-Tank General, and perform a thorough test on the proposed framework. The results show the effectiveness of our proposed framework and shed light on relevant issues on using the RL in RTS games.

Research paper thumbnail of An argument-based approach to reasoning with specificity

Artificial Intelligence, 2001

We present a new priority-based approach to reasoning with specificity which subsumes inheritance... more We present a new priority-based approach to reasoning with specificity which subsumes inheritance reasoning. The new approach differs from other priority-based approaches in the literature in the way priority between defaults is handled. Here, it is conditional rather than unconditional as in other approaches. We show that any unconditional handling of priorities between defaults as advocated in the literature until now is not sufficient to capture general defeasible inheritance reasoning. We propose a simple and novel argumentation semantics for reasoning with specificity taking the conditionality of the priorities between defaults into account. Since the proposed argumentation semantics is a form of stable semantics of nonmonotonic reasoning, it inherits a common problem of the latter where it is not always defined for every default theory. We propose a class of stratified default theories for which the argumentation semantics is always defined. We also show that acyclic and consistent inheritance networks are stratified. We prove that the argumentation semantics satisfies the basic properties of a nonmonotonic consequence relation such as deduction, reduction, conditioning, and cumulativity for well-defined and stratified default theories. We give a modular and polynomial transformation of default theories with specificity into semantically equivalent Reiter default theories.

Research paper thumbnail of Experiments with Online Reinforcement Learning in Real-Time Strategy Games

Applied Artificial Intelligence, 2009

Real-Time Strategy (RTS) games provide a challenging platform to implement online reinforcement l... more Real-Time Strategy (RTS) games provide a challenging platform to implement online reinforcement learning (RL) techniques in a real application. Computer as one player monitors opponents' (human or other computers) strategies and then updates its own policy using RL methods. In this paper, we firstly examine the suitability of applying the online RL in various computer games. RL application depends much on both RL complexity and the game features. We then propose a multi-layer framework for implementing online RL in an RTS game. The framework significantly reduces RL computational complexity by decomposing the state space in a hierarchical manner. We implement an RTS game-Tank General, and perform a thorough test on the proposed framework. We consider three typical profiles of RTS game players and compare two basic RL techniques applied in the game. The results show the effectiveness of our proposed framework and shed light on relevant issues on using online RL in RTS games.