Theory and Algorithms Commons^™

Understanding Sequential Decisions Via Inverse Reinforcement Learning, Siyuan Liu, Miguel Araujo, Emma Brunskill, Rosaldo Rossetti, Joao Barros, Ramayya Krishnan

Research Collection School Of Computing and Information Systems

The execution of an agent's complex activities, comprising sequences of simpler actions, sometimes leads to the clash of conflicting functions that must be optimized. These functions represent satisfaction, short-term as well as long-term objectives, costs and individual preferences. The way that these functions are weighted is usually unknown even to the decision maker. But if we were able to understand the individual motivations and compare such motivations among individuals, then we would be able to actively change the environment so as to increase satisfaction and/or improve performance. In this work, we approach the problem of providing highlevel and intelligible descriptions …

Iterative Statistical Verification Of Probabilistic Plans, Colin M. Potts

Lawrence University Honors Projects

Artificial intelligence seeks to create intelligent agents. An agent can be anything: an autopilot, a self-driving car, a robot, a person, or even an anti-virus system. While the current state-of-the-art may not achieve intelligence (a rather dubious thing to quantify) it certainly achieves a sense of autonomy. A key aspect of an autonomous system is its ability to maintain and guarantee safety—defined as avoiding some set of undesired outcomes. The piece of software responsible for this is called a planner, which is essentially an automated problem solver. An advantage computer planners have over humans is their ability to consider and …

Detecting Student Dropouts Using Fuzzy Inferencing, Shahriar Husainy

Theses and Dissertations

Fuzzy logic provides a methodology for reasoning using imprecise rules and assertions. Fuzzy inference is the process of formulating the mapping from a given input to an output using fuzzy logic. The mapping then provides a basis from which decisions can be made, or patterns discerned. This study concerns the development of a Fuzzy Inference System (FIS) for identifying likely student dropouts at Columbus State University (CSU). The fuzzy inference based model uses a hybrid knowledge extraction process to predict how likely each freshman student will be to drop their program of study at the end of their first semester. …

Real Time Digital Night Vision Using Nonlinear Contrast Enhancement, Nishikar Sapkota

UNLV Theses, Dissertations, Professional Papers, and Capstones

This thesis describes a nonlinear contrast enhancement technique to implement night vision in digital video. It is based on the global histogram equalization algorithm. First, the effectiveness of global histogram equalization is examined for images taken in low illumination environments in terms of Peak signal to noise ratio (PSNR) and visual inspection of images. Our analysis establishes the existence of an optimum intensity for which histogram equalization yields the best results in terms of output image quality in the context of night vision. Based on this observation, an incremental approach to histogram equalization is developed which gives better results than …

Scheduling Jobs On Two Uniform Parallel Machines To Minimize The Makespan, Sandhya Kodimala

UNLV Theses, Dissertations, Professional Papers, and Capstones

The problem of scheduling n independent jobs on m uniform parallel machines such that the total completion time is minimized is a NP-Hard problem. We propose several heuristic-based online algorithms for machines with different speeds called Q₂||C_max. To show the efficiency of the proposed online algorithms, we compute the optimal solution for Q₂||C_max using pseudo-polynomial algorithms based on dynamic programming method. The pseudo-polynomial algorithm has time complexity O (n T²) and can be run on reasonable time for small number of jobs and small processing times. This optimal offline algorithm is …

Calibration Of Traffic Flow Models, Victor Molano

College of Engineering: Graduate Celebration Programs

This study proposes a methodology to calibrate microscopic traffic flow simulation models. A Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm searches for the set of model parameters that minimizes the difference between actual and simulated values

Fast Sobel Edge Detection Using Parallel Pipeline-Based Architecture On Fpga, Mohammad Shokrolah Shirazi, Brendan Morris

College of Engineering: Graduate Celebration Programs

Implementing image processing algorithms on FPGA has recently become more popular since it provides high speed in comparison with software-based approaches. In this paper, we have presented fast pipeline-based architecture for one of the most popular edge detection algorithms called Sobel edge detection. The objective of our work is to present two fast pipeline-based architectures for Sobel edge detection on FPGA benefiting one and two way parallelism. We used Verilog language to implement our designs and we synthesized each one for Cyclone IV FPGA. Experimental results show that our pipeline-based architectures perform edge detection process more than 379 and 751 …

On High-Performance Parallel Decimal Fixed-Point Multiplier Designs, Ming Zhu

College of Engineering: Graduate Celebration Programs

Decimal computations are required in finance, and etc.

• Precise representation for decimals (E.g. 0.2, 0.7… )
• Performance Requirements (Software simulations are very slow)

Image Processing Algorithms For Improving Planetary Exploration And Understanding, Ali Pouryazdanpanah

College of Engineering: Graduate Celebration Programs

• To design a fully automated tool-set that allows to detect and extract the sky region in planetary images.

• To develop the new method for rock segmentation in planetary stereo images.
• To develop the new method for shadow detection in planetary images

Mono-Sized Sphere Packing Algorithm Development Using Optimized Monte Carlo Technique, Karn Soontrapa, Yitung Chen

College of Engineering: Graduate Celebration Programs

In this research, fuel cell catalyst layer was developed using the optimized sphere packing algorithm. An optimization technique named adaptive random search technique (ARSET) was employed in this packing algorithm. The ARSET algorithm will generate the initial location of spheres and allow them to move in the random direction with the variable moving distance, randomly selected from the sampling range (a), based on the Lennard–Jones potential and Morse potential of the current and new configuration. The solid fraction values obtained from this developed algorithm are in the range of 0.610–0.624 while the actual processing time can significantly be reduced by …

Operating System Scheduling: Linux Preemptive Scheduling Algorithms, Andrew Brand

Honors Capstones

Capstone submitted as a graduation requirement for the BSU Honors Program.

Artificial Immune Systems And Particle Swarm Optimization For Solutions To The General Adversarial Agents Problem, Jeremy Mange

Dissertations

The general adversarial agents problem is an abstract problem description touching on the fields of Artificial Intelligence, machine learning, decision theory, and game theory. The goal of the problem is, given one or more mobile agents, each identified as either “friendly" or “enemy", along with a specified environment state, to choose an action or series of actions from all possible valid choices for the next “timestep" or series thereof, in order to lead toward a specified outcome or set of outcomes. This dissertation explores approaches to this problem utilizing Artificial Immune Systems, Particle Swarm Optimization, and hybrid approaches, along with …

Impact Of Primary User Activity On The Performance Of Energy-Based Spectrum Sensing In Cognitive Radio Systems, Sara L. Macdonald

Electrical & Computer Engineering Theses & Dissertations

Increasing numbers of wireless devices and mobile data requirements have led to a spectrum shortage. However spectrum utilization percentages are often low due to the current static spectrum allocation process where primary users (PUs) are given exclusive use to spectrum. Several mechanisms to increase spectrum utilization have been proposed including opportunistic spectrum access (OSA). Cognitive Radio (CR) is an emerging concept in wireless communication systems that aims to enable OSA in licensed frequencies by secondary users (SUs). CR systems are expected to sense the spectrum in order to determine if the PU is transmitting. Therefore OSA performance relies on the …

Modeling Social Information Learning Among Taxi Drivers, Siyuan Liu, Ramayya Krishnan, Emma Brunskill, Lionel Ni

Research Collection School Of Computing and Information Systems

When a taxi driver of an unoccupied taxi is seeking passengers on a road unknown to him or her in a large city, what should the driver do? Alternatives include cruising around the road or waiting for a time period at the roadside in the hopes of finding a passenger or just leaving for another road enroute to a destination he knows (e.g., hotel taxi rank)? This is an interesting problem that arises everyday in many cities worldwide. There could be different answers to the question poised above, but one fundamental problem is how the driver learns about the likelihood …

Confidence Weighted Mean Reversion Strategy For Online Portfolio Selection, Bin Li, Steven C. H. Hoi, Peilin Zhao, Vivekanand Gopalkrishnan

Research Collection School Of Computing and Information Systems

Online portfolio selection has been attracting increasing attention from the data mining and machine learning communities. All existing online portfolio selection strategies focus on the first order information of a portfolio vector, though the second order information may also be beneficial to a strategy. Moreover, empirical evidence shows that relative stock prices may follow the mean reversion property, which has not been fully exploited by existing strategies. This article proposes a novel online portfolio selection strategy named Confidence Weighted Mean Reversion (CWMR). Inspired by the mean reversion principle in finance and confidence weighted online learning technique in machine learning, CWMR …

Fault-Tolerant Coverage In Dense Wireless Sensor Networks, Akshaye Dhawan, Magdalena Parks

Mathematics, Computer Science & Statistics Faculty Publications

In this paper, we present methods to detect and recover from sensor failure in dense wireless sensor networks. In order to extend the lifetime of a sensor network while maintaining coverage, a minimal subset of the deployed sensors are kept active while the other sensors can enter a low power sleep state. Several distributed algorithms for coverage have been proposed in the literature. Faults are of particular concern in coverage algorithms since sensors go into a sleep state in order to conserve battery until woken up by active sensors. If these active sensors were to fail, this could lead to …

Interpreting Individual Classifications Of Hierarchical Networks, Will Landecker, Michael David Thomure, Luis M.A. Bettencourt, Melanie Mitchell, Garrett T. Kenyon, Steven P. Brumby

Computer Science Faculty Publications and Presentations

Hierarchical networks are known to achieve high classification accuracy on difficult machine-learning tasks. For many applications, a clear explanation of why the data was classified a certain way is just as important as the classification itself. However, the complexity of hierarchical networks makes them ill-suited for existing explanation methods. We propose a new method, contribution propagation, that gives per-instance explanations of a trained network's classifications. We give theoretical foundations for the proposed method, and evaluate its correctness empirically. Finally, we use the resulting explanations to reveal unexpected behavior of networks that achieve high accuracy on visual object-recognition tasks using well-known …

Using Mapreduce Streaming For Distributed Life Simulation On The Cloud, Atanas Radenski

Mathematics, Physics, and Computer Science Faculty Books and Book Chapters

Distributed software simulations are indispensable in the study of large-scale life models but often require the use of technically complex lower-level distributed computing frameworks, such as MPI. We propose to overcome the complexity challenge by applying the emerging MapReduce (MR) model to distributed life simulations and by running such simulations on the cloud. Technically, we design optimized MR streaming algorithms for discrete and continuous versions of Conway’s life according to a general MR streaming pattern. We chose life because it is simple enough as a testbed for MR’s applicability to a-life simulations and general enough to make our results applicable …

Accelerated Data Delivery Architecture, Michael L. Grecol

College of Graduate Studies: Theses & Dissertations

This paper introduces the Accelerated Data Delivery Architecture (ADDA). ADDA establishes a framework to distribute transactional data and control consistency to achieve fast access to data, distributed scalability and non-blocking concurrency control by using a clean declarative interface. It is designed to be used with web-based business applications. This framework uses a combination of traditional Relational Database Management System (RDBMS) combined with a distributed Not Only SQL (NoSQL) database and a browser-based database. It uses a single physical and conceptual database schema designed for a standard RDBMS driven application. The design allows the architect to assign consistency levels to entities …

A Parallel Template For Implementing Filters For Biological Correlation Networks, Kathryn Dempsey Cooper, Vladimir Ufimtsev, Sanjukta Bhowmick, Hesham Ali

Interdisciplinary Informatics Faculty Publications

High throughput biological experiments are critical for their role in systems biology – the ability to survey the state of cellular mechanisms on the broad scale opens possibilities for the scientific researcher to understand how multiple components come together, and what goes wrong in disease states. However, the data returned from these experiments is massive and heterogeneous, and requires intuitive and clever computational algorithms for analysis. The correlation network model has been proposed as a tool for modeling and analysis of this high throughput data; structures within the model identified by graph theory have been found to represent key players …