Computer Sciences Commons^™

Decision Making Under Interval Uncertainty: Towards (Somewhat) More Convincing Justifications For Hurwicz Optimism-Pessimism Approach, Warattaya Chinnakum, Laura Berrout Ramos, Olugbenga Iyiola, Vladik Kreinovich

Departmental Technical Reports (CS)

In the ideal world, we know the exact consequences of each action. In this case, it is relatively straightforward to compare different possible actions and, as a result of this comparison, to select the best action. In real life, we only know the consequences with some uncertainty. A typical example is interval uncertainty, when we only know the lower and upper bounds on the expected gain. How can we compare such interval-valued alternatives? A usual way to compare such alternatives is to use the optimism-pessimism criterion developed by Nobelist Leo Hurwicz. In this approach, we maximize a weighted combination of …

Theoretical Explanation Of Recent Empirically Successful Code Quality Metrics, Vladik Kreinovich, Omar A. Masmali, Nguyen Hoang Phuong, Omar Badreddin

Departmental Technical Reports (CS)

Millions of lines of code are written every day, and it is not practically possible to perfectly thoroughly test all this code on all possible situations. In practice, we need to be able to separate codes which are more probable to contain bugs -- and which thus need to be tested more thoroughly -- from codes which are less probable to contain flaws. Several numerical characteristics -- known as code quality metrics -- have been proposed for this separation. Recently, a new efficient class of code quality metrics have been proposed, based on the idea to assign consequent integers to …

Quantum (And More General) Models Of Research Collaboration, Oscar Galindo, Miroslav Svitek, Vladik Kreinovich

Departmental Technical Reports (CS)

In the last decades, several papers have shown that quantum techniques can be successful in describing not only events in the micro-scale physical world -- for which they were originally invented -- but also in describing social phenomena, e.g., different economic processes. In our previous paper, we provide an explanation for this somewhat surprising successes. In this paper, we extend this explanation and show that quantum (and more general) techniques can also be used to model research collaboration.

Which Are The Correct Membership Functions? Correct "And"- And "Or"- Operations? Correct Defuzzification Procedure?, Olga Kosheleva, Vladik Kreinovich, Shahnaz Shahbazova

Departmental Technical Reports (CS)

Even in the 1990s, when many successful examples of fuzzy control appeared all the time, many users were somewhat reluctant to use fuzzy control. One of the main reasons for this reluctance was the perceived subjective character of fuzzy techniques -- for the same natural-language rules, different experts may select somewhat different membership functions and thus get somewhat different control/recommendation strategies. In this paper, we promote the idea that this selection does not have to be subjective. We can always select the "correct" membership functions, i.e., functions for which, on previously tested case, we got the best possible control. Similarly, …

Scale-Invariance And Fuzzy Techniques Explain The Empirical Success Of Inverse Distance Weighting And Of Dual Inverse Distance Weighting In Geosciences, Laxman Bokati, Aaron A. Velasco, Vladik Kreinovich

Departmental Technical Reports (CS)

Once we measure the values of a physical quantity at certain spatial locations, we need to interpolate these values to estimate the value of this quantity at other locations x. In geosciences, one of the most widely used interpolation techniques is inverse distance weighting, when we combine the available measurement results with the weights inverse proportional to some power of the distance from x to the measurement location. This empirical formula works well when measurement locations are uniformly distributed, but it leads to biased estimates otherwise. To decrease this bias, researchers recently proposed a more complex dual inverse distance weighting …

How To Combine (Dis)Utilities Of Different Aspects Into A Single (Dis)Utility Value, And How This Is Related To Geometric Images Of Happiness, Laxman Bokati, Nguyen Hoang Phuong, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

In many practical situations, a user needs our help in selecting the best out of a large number of alternatives. To be able to help, we need to understand the user's preferences. In decision theory, preferences are described by numerical values known as utilities. It is often not feasible to ask to user to provide utilities of all possible alternatives, so we must be able to estimate these utilities based on utilities of different aspects of these alternatives. In this paper, we provide a general formula for combining utilities of aspects into a single utility value. The resulting formula …

How To Describe Conditions Like 2-Out-Of-5 In Fuzzy Logic: A Neural Approach, Olga Kosheleva, Vladik Kreinovich, Nguyen Hoang Phuong

Departmental Technical Reports (CS)

In many medical applications, we diagnose a disease and/or apply a certain remedy if, e.g., two out of five conditions are satisfied. In the fuzzy case, i.e., when we only have certain degrees of confidence that each of n statement is satisfied, how do we estimate the degree of confidence that k out of n conditions are satisfied? In principle, we can get this estimate if we use the usual methodology of applying fuzzy techniques: we represent the desired statement in terms of "and" and "or", and use fuzzy analogues of these logical operations. The problem with this approach is …

How Quantum Cryptography And Quantum Computing Can Make Cyber-Physical Systems More Secure, Deepak Tosh, Oscar Galindo, Vladik Kreinovich, Olga Kosheleva

Departmental Technical Reports (CS)

For cyber-physical systems, cyber-security is vitally important. There are many cyber-security tools that make communications secure -- e.g., communications between sensors and the computers processing the sensor's data. Most of these tools, however, are based on RSA encryption, and it is known that with quantum computing, this encryption can be broken. It is therefore desirable to use an unbreakable alternative -- quantum cryptography -- for such communications. In this paper, we discuss possible consequences of this option. We also explain how quantum computers can help even more: namely, they can be used to optimize the system's design -- in particular, …

Why Squashing Functions In Multi-Layer Neural Networks, Julio Urenda, Orsoly Csiszár, Gábor Csiszár, József Dombi, Olga Kosheleva, Vladik Kreinovich, György Eigner

Departmental Technical Reports (CS)

Most multi-layer neural networks used in deep learning utilize rectified linear neurons. In our previous papers, we showed that if we want to use the exact same activation function for all the neurons, then the rectified linear function is indeed a reasonable choice. However, preliminary analysis shows that for some applications, it is more advantageous to use different activation functions for different neurons -- i.e., select a family of activation functions instead, and select the parameters of activation functions of different neurons during training. Specifically, this was shown for a special family of squashing functions that contain rectified linear neurons …

A Mystery Of Human Biological Development -- Can It Be Used To Speed Up Computations?, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

For many practical problems, the only known algorithms for solving them require non-feasible exponential time. To make computations feasible, we need an exponential speedup. A reasonable way to look for such possible speedup is to search for real-life phenomena where such a speedup can be observed. A natural place to look for such a speedup is to analyze the biological activities of human beings -- since we, after all, solve many complex problems that even modern super-fast computers have trouble solving. Up to now, this search was not successful -- e.g., there are people who compute much faster than others, …

How To Gauge The Quality Of A Testing Method When Ground Truth Is Known With Uncertainty, Nicholas Gray, Scott Ferson, Vladik Kreinovich

Departmental Technical Reports (CS)

The quality of a testing method is usually measured by using sensitivity, specificity, and/or precision. To compute each of these three characteristics, we need to know the ground truth, i.e., we need to know which objects actually have the tested property. In many applications (e.g., in medical diagnostics), the information about the objects comes from experts, and this information comes with uncertainty. In this paper, we show how to take this uncertainty into account when gauging the quality of testing methods.

Need For Simplicity And Everything Is A Matter Of Degree: How Zadeh's Philosophy Is Related To Kolmogorov Complexity, Quantum Physics, And Deep Learning, Vladik Kreinovich, Olga Kosheleva, Andres Ortiz-Muñoz

Departmental Technical Reports (CS)

Many people remember Lofti Zadeh's mantra -- that everything is a matter of degree. This was one of the main principles behind fuzzy logic. What is somewhat less remembered is that Zadeh also used another important principle -- that there is a need for simplicity. In this paper, we show that together, these two principles can generate the main ideas behind such various subjects as Kolmogorov complexity, quantum physics, and deep learning. We also show that these principles can help provide a better understanding of an important notion of space-time causality.

Using Prosody To Spot Location Mentions, Gerardo Cervantes

Open Access Theses & Dissertations

Identifying location mentions in speech is important for many information retrieval and information extraction tasks; here I explore the use of prosody for location spotting. While previous work has explored the use of prosody for spotting named entities, including locations, the specific value of prosody for finding locations in spontaneous speech has not been measured. Using the Switchboard corpus and LSTM modeling I obtain results indicating that prosody is useful in spotting location mentions. Further, I identify specific prosodic

features that tend to mark locations in American English.

A Comprehensive And Modular Robotic Control Framework For Model-Less Control Law Development Using Reinforcement Learning For Soft Robotics, Charles Sullivan

Open Access Theses & Dissertations

Soft robotics is a growing field in robotics research. Heavily inspired by biological systems, these robots are made of softer, non-linear, materials such as elastomers and are actuated using several novel methods, from fluidic actuation channels to shape changing materials such as electro-active polymers. Highly non-linear materials make modeling difficult, and sensors are still an area of active research. These issues have rendered typical control and modeling techniques often inadequate for soft robotics. Reinforcement learning is a branch of machine learning that focuses on model-less control by mapping states to actions that maximize a specific reward signal. Reinforcement learning has …

Abstraction Techniques In Security Games With Underlying Network Structure, Anjon Basak

Open Access Theses & Dissertations

In a multi-agent system, multiple intelligent agents interact with each other in an environment to achieve their objectives. They can do this because they know which actions are available to them and which actions they prefer to take in a particular situation. The job of game theory is to analyze the interactions of the intelligent agents by different solution techniques and provide analysis such as predicting outcomes or recommending courses of action to specific players. To do so game theory works with a model of real-world scenarios which helps us to make a better decision in our already complex daily …

Autonomous Trading Strategies For Dynamic Energy Markets, Moinul Morshed Porag Chowdhury

Open Access Theses & Dissertations

With increasing energy demand and an intermittent supply of renewable energy sources, our current energy grid needs a transformation towards a more robust, reliable energy trading architecture. The smart grid promises this architecture as the future of the present energy market, where traders will use digital technologies to automate the management of power delivery. It will improve many issues of the current energy grid such as sustainable, clean, renewable, reliable and secure energy supply, customer participation in markets, distributed generation, and transparency in energy trading. Using autonomous trading agents, we can bridge several dynamic energy markets and ensure an efficient …

Understanding The Digital Lives Of Transnational Students: A Case Study, Chowaing Chagra Belekeh

Open Access Theses & Dissertations

The proliferation and the fast-paced evolution of digital information communication technologies (ICTs) in contemporary times have arguably raised concern for us to comprehend what we do with these technologies and what these technologies do for us. The experience of engaging these technologies may not necessarily be the same for everyone—especially students who come from around the world to attain post-graduate degrees in the United States. This research focused on understanding the digital lives, choices, and experiences of transnational students who navigate and negotiate geopolitical borders and boundaries (physical)– in their quest for education. Using a case study analysis and collecting …

A Comparative Study Of The Impact Of Depth In Deep Learning Architectures, Kirsten Byers

Open Access Theses & Dissertations

Machine Learning continues to evolve as applications become more complex. Neural Networks, or Deep Networks, are integral to machine learning and the entire taxonomy of Artificial Intelligence [Sze17]. Intelligent structures and algorithms continue to advance, keeping pace with the complexi-ty of data. Changes in architecture, algorithms, and parameters are necessary to keep up with com-putational complexity and data available. This study focuses on how changes in depth of the archi-tecture affect performance on three distinct datasets, including one on Heart Disease. An adaptable network is created in original code, trained, and tested on these datasets. Its performance parameters are observed …

Compound Vision Approach For Autonomous Vehicles Navigation, Michael Mikhael

Open Access Theses & Dissertations

An analogy can be made between the sensing that occurs in simple robots and drones and that in insects and crustaceans, especially in basic navigation requirements. Thus, an approach in robots/drones based on compound eye vision could be useful. In this research, several image processing algorithms were used to detect and track moving objects starting with images upon which a grid (compound eye image) was superimposed, including contours detection, the second moments of those contours along with the grid applied to the original image, and Fourier Transforms and inverse Fourier Transforms. The latter also provide information about scene or camera …

Deep Learning For Overhead Imagery: Algorithms And Applications, Anthony Manuel Ortiz Cepeda

Open Access Theses & Dissertations

Remote sensing using overhead imagery has critical impact to the way we understand our environment and offers crucial information for scene understanding, climate change research, disaster response, urban planning, forest management, and many other applications. At present, deep learning is increasingly used in remote sensing, but mostly borrowing algorithms developed for natural images in the computer vision community. Specific challenges arise while applying deep learning to remote sensing. These challenges include issues related to the high dimensionality and limited labeled data, security and robustness to adversarial attacks, and model generalization. In this Thesis we focus on tackling these key challenges. …

Comparing Predictive Performance Of Statistical Learning Models On Medical Data, Francis Biney

Open Access Theses & Dissertations

This work investigates the predictive performance of 10 Machine learning models on three medical data including Breast cancer, Heart disease and Prostate cancer. Furthermore, we use the models to identify risk factors that contribute significantly to these diseases.

The models considered include; Logistic regression with L1 and L_2 penalties, Principal component logistic regression(PCR-LR), Partial least squares logistic regression(PLS-LR), Multivariate adaptive regression splines(MARS), Support vector machine with Radial Basis Kernel (SVM-RBK), Random Forest(RF), Gradient Boosting Machines(GBM), Elastic Net (Enet) and Feedforward Neural Network(FFNN). The models were grouped according to their similarities and learning style; i) Linear regularized models: LR-Lasso, LR-Ridge and …

Finalcache: Eviction Based On Implicit Entry Reachability, Adrian Veliz

Open Access Theses & Dissertations

Software caches for garbage collected systems are unaware which cache entries are referenced by a client program. Lacking this key information, software caches cannot determine the cache’s contribution to the program’s memory footprint. Furthermore, their eviction heuristics must rely on access history as a proxy for usage. Divergence between usage and access history can undermine the intention of eviction thereby resulting in problematic cache behavior.

This dissertation proposes designs for a novel family of “usage-based” software cache informed of entry reachability by the automatic memory management system. Unlike extant software caches, usage-based caches can accurately determine their memory footprint because …

Glacier Segmentation In Satellite Images For Hindu Kush Himalaya Region, Bibek Aryal

Open Access Theses & Dissertations

Climate change poses a risk to individuals whose livelihoods depend on the health of glacier ecosystems. Monitoring glaciers in the Himalayan Hindu Kush (HKH) region is of high importance especially when we consider the impact of recent climate change on them. Our work aims to provide an automated method to outline glaciers using machine learning techniques and publicly available remote sensing imagery.In this work, we present ways to delineate glaciers from Landsat-7 imagery using various machine learning and computer vision techniques. The multi-step methodology that we present in this work is generalizable across different types of satellite and overhead imagery, …

Towards The Development Of A Cohesive Design-Driven Code Quality Metrics, Omar Masmali

Open Access Theses & Dissertations

Software complexity is an indicator of expected future maintenance and sustainability. Excessive complexity suggests that software or a component of software has a design or implementation that is difficult to understand, modify, and maintain. Several complexity measures have been developed by researchers to identify and characterize degrees of complexity. Code smells are widely adopted as indicators for low code quality. Many studies have adopted fixed threshold values for code smells and other quality metrics. These fixed threshold values often ignore the uniqueness of each software system and the unique roles each component play. Moreover, these thresholds are largely fixed throughout …

Benchmarking Machine Learning Methods For Molecular Property Prediction, Govinda Bahadur Kc

Open Access Theses & Dissertations

Machine learning (ML) techniques have been widely applied in a variety of areas ranging from pattern recognition, natural language processing, and computer games to self-driving cars, clinical diagnostics, and molecular structure prediction easing day to day life of human beings. Drug discovery is an expensive, complex, and time taking process. Currently, the pharma industry is hoping to leverage machine learning methods in expediting the drug discovery process. Molecular property prediction is one of the most important tasks in drug discovery. While developing a new drug relies on a proper understanding of molecular properties, there has been great interest in the …

Brian Valdez - Dynamics And Control Of A 3-Dof Manipulator With Deep Learning Feedback, Brian Orlando Valdez

Open Access Theses & Dissertations

With the ever-increasing demands in the space domain and accessibility to low-cost small satellite platforms for educational and scientific projects, efforts are being made in various technology capacities including robotics and artificial intelligence in microgravity. The MIRO Center for Space Exploration and Technology Research (cSETR) prepares the development of their second nanosatellite to launch to space and it is with that opportunity that a 3-DOF robotic arm is in development to be one of the payloads in the nanosatellite. Analyses, hardware implementation, and testing demonstrate a potential positive outcome from including the payload in the nanosatellite and a deep learning …

Why Spiking Neural Networks Are Efficient: A Theorem, Michael Beer, Julio Urenda, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

Current artificial neural networks are very successful in many machine learning applications, but in some cases they still lag behind human abilities. To improve their performance, a natural idea is to simulate features of biological neurons which are not yet implemented in machine learning. One of such features is the fact that in biological neural networks, signals are represented by a train of spikes. Researchers have tried adding this spikiness to machine learning and indeed got very good results, especially when processing time series (and, more generally, spatio-temporal data). In this paper, we provide a theoretical explanation for this empirical …

Joule's 19th Century Energy Conservation Meta-Law And The 20th Century Physics (Quantum Mechanics And General Relativity): 21st Century Analysis, Vladik Kreinovich, Olga Kosheleva

Departmental Technical Reports (CS)

Joule's Energy Conservation Law was the first "meta-law": a general principle that all physical equations must satisfy. It has led to many important and useful physical discoveries. However, a recent analysis seems to indicate that this meta-law is inconsistent with other principles -- such as the existence of free will. We show that this conclusion about inconsistency is based on a seemingly reasonable -- but simplified -- analysis of the situation. We also show that a more detailed mathematical and physical analysis of the situation reveals that not only Joule's principle remains true -- it is actually strengthened: it is …

Why Gamma Distribution Of Seismic Inter-Event Times: A Theoretical Explanation, Laxman Bokati, Aaron A. Velasco, Vladik Kreinovich

Departmental Technical Reports (CS)

It is known that the distribution of seismic inter-event times is well described by the Gamma distribution. Recently, this fact has been used to successfully predict major seismic events. In this paper, we explain that the Gamma distribution of seismic inter-event times can be naturally derived from the first principles.

Finitely Generated Sets Of Fuzzy Values: If "And" Is Exact, Then "Or" Is Almost Always Approximate, And Vice Versa -- A Theorem, Julio Urenda, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

In the traditional fuzzy logic, experts' degrees of confidence are described by numbers from the interval [0,1]. Clearly, not all the numbers from this interval are needed: in the whole history of the Universe, there will be only countably many statements and thus, only countably many possible degree, while the interval [0,1] is uncountable. It is therefore interesting to analyze what is the set S of actually used values. The answer depends on the choice of "and"-operations (t-norms) and "or"-operations (t-conorms). For the simplest pair of min and max, any finite set will do -- as long as it is …