Computer Sciences Commons^™

Generic Online Learning For Partial Visible & Dynamic Environment With Delayed Feedback, Behrooz Shahriari

Master's Projects

Reinforcement learning (RL) has been applied to robotics and many other domains which a system must learn in real-time and interact with a dynamic environment. In most studies the state- action space that is the key part of RL is predefined. Integration of RL with deep learning method has however taken a tremendous leap forward to solve novel challenging problems such as mastering a board game of Go. The surrounding environment to the agent may not be fully visible, the environment can change over time, and the feedbacks that agent receives for its actions can have a fluctuating delay. In …

An Efficient Approach To Model-Based Hierarchical Reinforcement Learning, Zhuoru Li, Akshay Narayan, Tze-Yun Leong

Research Collection School Of Computing and Information Systems

We propose a model-based approach to hierarchical reinforcement learning that exploits shared knowledge and selective execution at different levels of abstraction, to efficiently solve large, complex problems. Our framework adopts a new transition dynamics learning algorithm that identifies the common action-feature combinations of the subtasks, and evaluates the subtask execution choices through simulation. The framework is sample efficient, and tolerates uncertain and incomplete problem characterization of the subtasks. We test the framework on common benchmark problems and complex simulated robotic environments. It compares favorably against the stateof-the-art algorithms, and scales well in very large problems.

Seapot-Rl: Selective Exploration Algorithm For Policy Transfer In Rl, Akshay Narayan, Zhuoru Li, Tze-Yun Leong

Research Collection School Of Computing and Information Systems

We propose a new method for transferring a policy from a source task to a target task in model-based reinforcement learning. Our work is motivated by scenarios where a robotic agent operates in similar but challenging environments, such as hospital wards, differentiated by structural arrangements or obstacles, such as furniture. We address problems that require fast responses adapted from incomplete, prior knowledge of the agent in new scenarios. We present an efficient selective exploration strategy that maximally reuses the source task policy. Reuse efficiency is effected through identifying sub-spaces that are different in the target environment, thus limiting the exploration …

Ai Education: Machine Learning Resources, Todd W. Neller

Computer Science Faculty Publications

In this column, we focus on resources for learning and teaching three broad categories of machine learning (ML): supervised, unsupervised, and reinforcement learning. In ournext column, we will focus specifically on deep neural network learning resources, so if you have any resource recommendations, please email them to the address above. [excerpt]

Towards Autonomous Behavior Learning Of Non-Player Characters In Games, Shu Feng, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Non-Player-Characters (NPCs), as found in computer games, can be modelled as intelligent systems, which serve to improve the interactivity and playability of the games. Although reinforcement learning (RL) has been a promising approach to creating the behavior models of non-player characters (NPC), an initial stage of exploration and low performance is typically required. On the other hand, imitative learning (IL) is an effective approach to pre-building a NPC’s behavior model by observing the opponent’s actions, but learning by imitation limits the agent’s performance to that of its opponents. In view of their complementary strengths, this paper proposes a computational model …

Reinforcement Learning Framework For Modeling Spatial Sequential Decisions Under Uncertainty: (Extended Abstract), Truc Viet Le, Siyuan Liu, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

We consider the problem of trajectory prediction, where a trajectory is an ordered sequence of location visits and corresponding timestamps. The problem arises when an agent makes sequential decisions to visit a set of spatial locations of interest. Each location bears a stochastic utility and the agent has a limited budget to spend. Given the agent's observed partial trajectory, our goal is to predict the remaining trajectory. We propose a solution framework to the problem considering both the uncertainty of utility and the budget constraint. We use reinforcement learning (RL) to model the underlying decision processes and inverse RL to …

Reinforcement Learning-Based Mobile Robot Navigation, Ni̇hal Altuntaş, Erkan İmal, Nahi̇t Emanet, Ceyda Nur Öztürk

Turkish Journal of Electrical Engineering and Computer Sciences

In recent decades, reinforcement learning (RL) has been widely used in different research fields ranging from psychology to computer science. The unfeasibility of sampling all possibilities for continuous-state problems and the absence of an explicit teacher make RL algorithms preferable for supervised learning in the machine learning area, as the optimal control problem has become a popular subject of research. In this study, a system is proposed to solve mobile robot navigation by opting for the most popular two RL algorithms, Sarsa($\lambda )$ and Q($\lambda )$. The proposed system, developed in MATLAB, uses state and action sets, defined in a …

Adaptive Duty Cycling In Sensor Networks With Energy Harvesting Using Continuous-Time Markov Chain And Fluid Models, Ronald Wai Hong Chan, Pengfei Zhang, Ido Nevat, Sai Ganesh Nagarajan, Alvin Cerdena Valera, Hwee Xian Tan

Research Collection School Of Computing and Information Systems

The dynamic and unpredictable nature of energy harvesting sources available for wireless sensor networks, and the time variation in network statistics like packet transmission rates and link qualities, necessitate the use of adaptive duty cycling techniques. Such adaptive control allows sensor nodes to achieve long-run energy neutrality, where energy supply and demand are balanced in a dynamic environment such that the nodes function continuously. In this paper, we develop a new framework enabling an adaptive duty cycling scheme for sensor networks that takes into account the node battery level, ambient energy that can be harvested, and application-level QoS requirements. We …

A Comparative Study Between Motivated Learning And Reinforcement Learning, James T. Graham, Janusz A. Starzyk, Zhen Ni, Haibo He, T.-H. Teng, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

This paper analyzes advanced reinforcement learning techniques and compares some of them to motivated learning. Motivated learning is briefly discussed indicating its relation to reinforcement learning. A black box scenario for comparative analysis of learning efficiency in autonomous agents is developed and described. This is used to analyze selected algorithms. Reported results demonstrate that in the selected category of problems, motivated learning outperformed all reinforcement learning algorithms we compared with.

Multiagent-Based Simulation Of Simultaneous Electricity Market Auctions In Restructured Environment, Mohammad Farshad

Turkish Journal of Electrical Engineering and Computer Sciences

In the restructured environment of the power industry, various commodities such as energy and operating reserves may be provided through simultaneous auctions. Prediction of market players' behavior in the auctions and simulation of the markets' environment can assist market decision-makers in evaluating specific policies before enforcing them in the real environment. Considering effects of the energy and varieties of reserve markets and also their interactions in the simulations is of high importance, which leads to more realistic simulation results. In this paper, an approach based on a multiagent system is proposed for simulating the simultaneous energy, spinning reserve, and replacement …

Reinforcement Learning Of Distributed Surveillance Plans, Madhavi Chittireddy

Master's Theses

This thesis describes the design and implementation of a Reinforcement Learning algorithm on a camera surveillance model which is used to know the stackelberg strategies of attacker and defender. This reinforcement learning algorithm is compared with the uniform policy and hill climbing algorithms by executing them on a common set of different data files, generated programmatically with various combinations of problem size, location, and orientation transitions as well as rewards of attacker and defender. The comparison includes the time taken to obtain better stackelberg policy and the resulted final pay-off of the defender. This thesis shows that the reinforcement learning …

Convergence Of A Reinforcement Learning Algorithm In Continuous Domains, Stephen Carden

All Dissertations

In the field of Reinforcement Learning, Markov Decision Processes with a finite number of states and actions have been well studied, and there exist algorithms capable of producing a sequence of policies which converge to an optimal policy with probability one. Convergence guarantees for problems with continuous states also exist. Until recently, no online algorithm for continuous states and continuous actions has been proven to produce optimal policies. This Dissertation contains the results of research into reinforcement learning algorithms for problems in which both the state and action spaces are continuous. The problems to be solved are introduced formally as …

Integrating Motivated Learning And K-Winner-Take-All To Coordinate Multi-Agent Reinforcement Learning, Teck-Hou Teng, Ah-Hwee Tan, Janusz Starzyk, Yuan-Sin Tan, Loo-Nin Teow

Research Collection School Of Computing and Information Systems

This work addresses the coordination issue in distributed optimization problem (DOP) where multiple distinct and time-critical tasks are performed to satisfy a global objective function. The performance of these tasks has to be coordinated due to the sharing of consumable resources and the dependency on non-consumable resources. Knowing that it can be sub-optimal to predefine the performance of the tasks for large DOPs, the multi-agent reinforcement learning (MARL) framework is adopted wherein an agent is used to learn the performance of each distinct task using reinforcement learning. To coordinate MARL, we propose a novel coordination strategy integrating Motivated Learning (ML) …

Creating Autonomous Adaptive Agents In A Real-Time First-Person Shooter Computer Game, Di Wang, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Games are good test-beds to evaluate AI methodologies. In recent years, there has been a vast amount of research dealing with real-time computer games other than the traditional board games or card games. This paper illustrates how we create agents by employing FALCON, a self-organizing neural network that performs reinforcement learning, to play a well-known first-person shooter computer game called Unreal Tournament. Rewards used for learning are either obtained from the game environment or estimated using the temporal difference learning scheme. In this way, the agents are able to acquire proper strategies and discover the effectiveness of different weapons without …

Memory-Guided Exploration In Reinforcement Learning, James L. Carroll, Todd Peterson

Journal of Undergraduate Research

Traditional reinforcement learning techniques learn a single task by giving the agent positive and negative rewards. In one type of reinforcement learning, called Q-learning, the agent stores Qvalues, which are the expected reward for performing an action in a given state. Task transfer is a method of transferring information learned in one task to another related task. Most work in transfer has focused on classification techniques. The purpose of our research has been to extend classification techniques to reinforcement learning.

Reinforcement Learning Task Clustering, James Carroll, Todd Peterson

Journal of Undergraduate Research

Reinforcement Learning is a process whereby actions are acquired using reinforcement signals. A signal is given to an autonomous agent indicating how well that agent is performing an action. The agent then attempts to maximize this reinforcement signal. One common method in reinforcement learning is Q-learning where the agent attempts to learn the expected temporally discounted value function for performing an action a in a state s Q(s,a). This function is updated according to:

Complementary Layered Learning, Sean Mondesire

Electronic Theses and Dissertations

Layered learning is a machine learning paradigm used to develop autonomous robotic-based agents by decomposing a complex task into simpler subtasks and learns each sequentially. Although the paradigm continues to have success in multiple domains, performance can be unexpectedly unsatisfactory. Using Boolean-logic problems and autonomous agent navigation, we show poor performance is due to the learner forgetting how to perform earlier learned subtasks too quickly (favoring plasticity) or having difficulty learning new things (favoring stability). We demonstrate that this imbalance can hinder learning so that task performance is no better than that of a suboptimal learning technique, monolithic learning, which …

Adaptive Computer‐Generated Forces For Simulator‐Based Training, Expert Systems With Applications, Teck-Hou Teng, Ah-Hwee Tan, Loo-Nin Teow

Research Collection School Of Computing and Information Systems

Simulator-based training is in constant pursuit of increasing level of realism. The transition from doctrine-driven computer-generated forces (CGF) to adaptive CGF represents one such effort. The use of doctrine-driven CGF is fraught with challenges such as modeling of complex expert knowledge and adapting to the trainees’ progress in real time. Therefore, this paper reports on how the use of adaptive CGF can overcome these challenges. Using a self-organizing neural network to implement the adaptive CGF, air combat maneuvering strategies are learned incrementally and generalized in real time. The state space and action space are extracted from the same hierarchical doctrine …

Actor-Critic-Based Ink Drop Spread As An Intelligent Controller, Hesam Sagha, Iman Esmaili Paeen Afrakoti, Saeed Bagherishouraki

Turkish Journal of Electrical Engineering and Computer Sciences

This paper introduces an innovative adaptive controller based on the actor-critic method. The proposed approach employs the ink drop spread (IDS) method as its main engine. The IDS method is a new trend in soft-computing approaches that is a universal fuzzy modeling technique and has been also used as a supervised controller. Its process is very similar to the processing system of the human brain. The proposed actor-critic method uses an IDS structure as an actor and a 2-dimensional plane, representing control variable states, as a critic that estimates the lifetime goodness of each state. This method is fast, simple, …

Self-Regulating Action Exploration In Reinforcement Learning, Teck-Hou Teng, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

The basic tenet of a learning process is for an agent to learn for only as much and as long as it is necessary. With reinforcement learning, the learning process is divided between exploration and exploitation. Given the complexity of the problem domain and the randomness of the learning process, the exact duration of the reinforcement learning process can never be known with certainty. Using an inaccurate number of training iterations leads either to the non-convergence or the over-training of the learning agent. This work addresses such issues by proposing a technique to self-regulate the exploration rate and training duration …

Efficient Reinforcement Learning In Multiple-Agent Systems And Its Application In Cognitive Radio Networks, Jing Zhang

Dissertations

The objective of reinforcement learning in multiple-agent systems is to find an efficient learning method for the agents to behave optimally. Finding Nash equilibrium has become the common learning target for the optimality. However, finding Nash equilibrium is a PPAD (Polynomial Parity Arguments on Directed graphs)-complete problem. The conventional methods can find Nash equilibrium for some special types of Markov games.

This dissertation proposes a new reinforcement learning algorithm to improve the search efficiency and effectiveness for multiple-agent systems. This algorithm is based on the definition of Nash equilibrium and utilizes the greedy and rational features of the agents. When …

Motivated Learning For The Development Of Autonomous Agents, Janusz A. Starzyk, James T. Graham, Pawel Raif, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

A new machine learning approach known as motivated learning (ML) is presented in this work. Motivated learning drives a machine to develop abstract motivations and choose its own goals. ML also provides a self-organizing system that controls a machine’s behavior based on competition between dynamically-changing pain signals. This provides an interplay of externally driven and internally generated control signals. It is demonstrated that ML not only yields a more sophisticated learning mechanism and system of values than reinforcement learning (RL), but is also more efficient in learning complex relations and delivers better performance than RL in dynamically changing environments. In …

Self‐Regulating Action Exploration In Reinforcement Learning, Teck-Hou Teng, Ah-Hwee Tan, Yuan-Sin Tan

Research Collection School Of Computing and Information Systems

The basic tenet of a learning process is for an agent to learn for only as much and as long as it is necessary. With reinforcement learning, the learning process is divided between exploration and exploitation. Given the complexity of the problem domain and the randomness of the learning process, the exact duration of the reinforcement learning process can never be known with certainty. Using an inaccurate number of training iterations leads either to the non-convergence or the over-training of the learning agent. This work addresses such issues by proposing a technique to self-regulate the exploration rate and training duration …

Cooperative Reinforcement Learning In Topology-Based Multi-Agent Systems, Dan Xiao, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Topology-based multi-agent systems (TMAS), wherein agents interact with one another according to their spatial relationship in a network, are well suited for problems with topological constraints. In a TMAS system, however, each agent may have a different state space, which can be rather large. Consequently, traditional approaches to multi-agent cooperative learning may not be able to scale up with the complexity of the network topology. In this paper, we propose a cooperative learning strategy, under which autonomous agents are assembled in a binary tree formation (BTF). By constraining the interaction between agents, we effectively unify the state space of individual …

A Hybrid Agent Architecture Integrating Desire, Intention And Reinforcement Learning, Ah-Hwee Tan, Yew-Soon Ong, Akejariyawong Tapanuj

Research Collection School Of Computing and Information Systems

This paper presents a hybrid agent architecture that integrates the behaviours of BDI agents, specifically desire and intention, with a neural network based reinforcement learner known as Temporal DifferenceFusion Architecture for Learning and COgNition (TD-FALCON). With the explicit maintenance of goals, the agent performs reinforcement learning with the awareness of its objectives instead of relying on external reinforcement signals. More importantly, the intention module equips the hybrid architecture with deliberative planning capabilities, enabling the agent to purposefully maintain an agenda of actions to perform and reducing the need of constantly sensing the environment. Through reinforcement learning, plans can also be …

Higher-Level Application Of Adaptive Dynamic Programming/Reinforcement Learning – A Next Phase For Controls And System Identification?, George G. Lendaris

Systems Science Friday Noon Seminar Series

Humans have the ability to make use of experience while performing system identification and selecting control actions for changing situations. In contrast to current technological implementations that slow down as more knowledge is stored, as more experience is gained, human processing speeds up and has enhanced effectiveness. An emerging experience-based (“higher level”) approach promises to endow our technology with enhanced efficiency and effectiveness.

The notions of context and context discernment are important to understanding this human ability. These are defined as appropriate to controls and system-identification. Some general background on controls, Dynamic Programming, and Adaptive Critic leading to Adaptive Dynamic …

An Exploration Of Multi-Agent Learning Within The Game Of Sheephead, Brady Brau

All Graduate Theses, Dissertations, and Other Capstone Projects

In this paper, we examine a machine learning technique presented by Ishii et al. used to allow for learning in a multi-agent environment and apply an adaptation of this learning technique to the card game Sheephead. We then evaluate the effectiveness of our adaptation by running simulations against rule-based opponents. Multi-agent learning presents several layers of complexity on top of a single-agent learning in a stationary environment. This added complexity and increased state space is just beginning to be addressed by researchers. We utilize techniques used by Ishii et al. to facilitate this multi-agent learning. We model the environment of …

Reinforcement Learning Of Competitive And Cooperative Skills In Soccer Agents, Jinsong Leng, Chee Lim

Research outputs 2011

The main aim of this paper is to provide a comprehensive numerical analysis on the efficiency of various reinforcementlearning (RL) techniques in an agent-based soccer game. The SoccerBots is employed as a simulation testbed to analyze the effectiveness of RL techniques under various scenarios. A hybrid agent teaming framework for investigating agent team architecture, learning abilities, and other specific behaviours is presented. Novel RL algorithms to verify the competitiveandcooperativelearning abilities of goal-oriented agents for decision-making are developed. In particular, the tile coding (TC) technique, a function approximation approach, is used to prevent the state space from growing exponentially, hence avoiding …

A Biologically-Inspired Cognitive Agent Model Integrating Declarative Knowledge And Reinforcement Learning, Ah-Hwee Tan, Gee-Wah Ng

Research Collection School Of Computing and Information Systems

The paper proposes a biologically-inspired cognitive agent model, known as FALCON-X, based on an integration of the Adaptive Control of Thought (ACT-R) architecture and a class of self-organizing neural networks called fusion Adaptive Resonance Theory (fusion ART). By replacing the production system of ACT-R by a fusion ART model, FALCON-X integrates high-level deliberative cognitive behaviors and real-time learning abilities, based on biologically plausible neural pathways. We illustrate how FALCON-X, consisting of a core inference area interacting with the associated intentional, declarative, perceptual, motor and critic memory modules, can be used to build virtual robots for battles in a simulated RoboCode …

A Self-Organizing Neural Architecture Integrating Desire, Intention And Reinforcement Learning, Ah-Hwee Tan, Yu-Hong Feng, Yew-Soon Ong

Research Collection School Of Computing and Information Systems

This paper presents a self-organizing neural architecture that integrates the features of belief, desire, and intention (BDI) systems with reinforcement learning. Based on fusion Adaptive Resonance Theory (fusion ART), the proposed architecture provides a unified treatment for both intentional and reactive cognitive functionalities. Operating with a sense-act-learn paradigm, the low level reactive module is a fusion ART network that learns action and value policies across the sensory, motor, and feedback channels. During performance, the actions executed by the reactive module are tracked by a high level intention module (also a fusion ART network) that learns to associate sequences of actions …