Browsing by Subject "Learning"

Now showing 1 - 20 of 29

Open Access
A car following model with an attention-based cognitive framework: theory, application, and statistical analysis
(2024-08) Habboush, Seymanur Al
Traffic simulators are essential for testing autonomous driving algorithms, and they require driver models that accurately emulate human behavior to reflect real traffic conditions. This thesis focuses on developing human driver models to be used in these simulators. The limitations of fixed driver models, which do not adapt to new information, are addressed by introducing an attention-based learning mechanism inspired by human memory. This mechanism is integrated into a multi-type car-following model developed in this study. Unlike existing models, this approach allows the ego driver’s decisions to be influenced by both the vehicle in front and the driver behind them. The predictive capabilities of the proposed model is demonstrated using human driving data and a comprehensive statistical analysis of the model parameter distributions is provided. This analysis shows how the model captures general behavioral tendencies across different data sets, enhancing the understanding of interactions between human drivers and providing more realistic simulations for testing purposes. Finally, a step by step guide for using the model in the development of high-fidelity traffic models is presented.
Open Access
Aging wireless bandits: regret analysis and order-optimal learning algorithm
(IEEE, 2021-11-13) Atay, Eray Unsal; Kadota, Igor; Modiano, Eytan
We consider a single-hop wireless network with sources transmitting time-sensitive information to the destination over multiple unreliable channels. Packets from each source are generated according to a stochastic process with known statistics and the state of each wireless channel (ON/OFF) varies according to a stochastic process with unknown statistics. The reliability of the wireless channels is to be learned through observation. At every time-slot, the learning algorithm selects a single pair (source, channel) and the selected source attempts to transmit its packet via the selected channel. The probability of a successful transmission to the destination depends on the reliability of the selected channel. The goal of the learning algorithm is to minimize the Age-of-Information (AoI) in the network over T time-slots. To analyze its performance, we introduce the notion of AoI-regret, which is the difference between the expected cumulative AoI of the learning algorithm under consideration and the expected cumulative AoI of a genie algorithm that knows the reliability of the channels a priori. The AoI-regret captures the penalty incurred by having to learn the statistics of the channels over the T time-slots. The results are two-fold: first, we consider learning algorithms that employ well-known solutions to the stochastic multi-armed bandit problem (such as ϵ-Greedy, Upper Confidence Bound, and Thompson Sampling) and show that their AoI-regret scales as Θ(log T); second, we develop a novel learning algorithm and show that it has O(1) regret. To the best of our knowledge, this is the first learning algorithm with bounded AoI-regret.
Open Access
Announcements and credibility under inflation targeting
(Elsevier BV, 2008) Demir, B.; Yigit, T. M.
We inspect how inflation target announcements are instrumental in building central bank credibility and shaping inflation expectations. Investigating the role of announcements by using a time varying credibility measure, we find that both the accuracy and the frequency of inflation announcements have a positive impact on how much attention the public pays to target announcements. © 2008 Elsevier B.V. All rights reserved.
Open Access
Beyond hierarchies: emerging organizational structures that are non-hierarchical
(1995) Karpuzoğlu, Gürhan
Open Access
Categorization in a hierarchically structured text database
(2001) Kutlu, Ferhat
Over the past two decades there has been a huge increase in the amount of data being stored in databases and the on-line ﬂow of data by the effects of improvements in Internet. This huge increase brought out the needs for intelligent tools to manage that size of data and its ﬂow. Hierarchical approach is the best way to satisfy these needs and it is so widespread among people dealing with databases and Internet. Usenet newsgroups system is one of the on-line databases that have built-in hierarchical structures. Our point of departure is this hierarchical structure which makes categorization tasks easier and faster. In fact most of the search engines in Internet also exploit inherent hierarchy of Internet. Growing size of data makes most of the traditional categorization algorithms obsolete. Thus we developed a brand-new categorization learning algorithm which constructs an index tree out of Usenet news database and then decides the related newsgroups of a new news by categorizing it over the index tree. In learning phase it has an agglomerative and bottom-up hierarchical approach. In categorization phase it does an overlapping and supervised categorization. k Nearest Neighbor categorization algorithm is used to compare the complexity measure and accuracy of our algorithm. This comparison does not only mean comparing two different algorithms but also means comparing hierarchical approach vs. ﬂat approach, similarity measure vs. distance measure and importance of accuracy vs. importance of speed. Our algorithm prefers hierarchical approach and similarity measure, and greatly outperforms k Nearest Neighbor categorization algorithm in speed with minimal loss of accuracy.
Open Access
Contextual learning for unit commitment with renewable energy sources
(IEEE, 2017) Lee, H. -S.; Tekin, Cem; Schaar, M.; Lee, J. -W.
In this paper, we study a unit commitment (UC) problem minimizing operating costs of the power system with renewable energy sources. We develop a contextual learning algorithm for UC (CLUC) which learns which UC schedule to choose based on the context information such as past load demand and weather condition. CLUC does not require any prior knowledge on the uncertainties such as the load demand and the renewable power outputs, and learns them over time using the context information. We characterize the performance of CLUC analytically, and prove its optimality in terms of the long-term average cost. Through the simulation results, we show the performance of CLUC and the effectiveness of utilizing the context information in the UC problem.
Open Access
Developing empathy towards older adults in design
(Routledge, 2017-02-06) Altay, Burçak
In design disciplines, an affective understanding of users’ everyday lives can increase designer sensitivity and awareness, leading to higher-quality design outcomes. Developing students’ empathic understanding within design education is required to accomplish this goal. This article discusses learning strategies that enhance students’ empathic horizons, and specifically analyzes an assignment conducted in an Interior Architecture and Environmental Design course, “The Grandparent Experience.” Here, exposure through observation and interviewing, and art-based methods are employed to develop students’ empathy towards older adults. We conducted a survey with students who completed the exercise and the course, exploring their perspectives on their learning. The results reveal that students had positive views on the assignment’s effectiveness regarding the learning outcome and learning process. Implications for empathic design education and educational gerontology are discussed.
Restricted
Discursive practices and psychological science
(1992) Greenwood, John D.
Open Access
Dopamine replacement therapy, learning and reward prediction in Parkinson’s disease: Implications for rehabilitation
(Frontiers Research Foundation, 2016) Ferrazzoli, D.; Carter, A.; Ustun, F. S.; Palamara, G.; Ortelli, P.; Maestri, R.; Yucel, M.; Frazzitta, G.
The principal feature of Parkinson’s disease (PD) is the impaired ability to acquire and express habitual-automatic actions due to the loss of dopamine in the dorsolateral striatum, the region of the basal ganglia associated with the control of habitual behavior. Dopamine replacement therapy (DRT) compensates for the lack of dopamine, representing the standard treatment for different motor symptoms of PD (such as rigidity, bradykinesia and resting tremor). On the other hand, rehabilitation treatments, exploiting the use of cognitive strategies, feedbacks and external cues, permit to “learn to bypass” the defective basal ganglia (using the dorsolateral area of the prefrontal cortex) allowing the patients to perform correct movements under executive-volitional control. Therefore, DRT and rehabilitation seem to be two complementary and synergistic approaches. Learning and reward are central in rehabilitation: both of these mechanisms are the basis for the success of any rehabilitative treatment. Anyway, it is known that “learning resources” and reward could be negatively influenced from dopaminergic drugs. Furthermore, DRT causes different well-known complications: among these, dyskinesias, motor fluctuations, and dopamine dysregulation syndrome (DDS) are intimately linked with the alteration in the learning and reward mechanisms and could impact seriously on the rehabilitative outcomes. These considerations highlight the need for careful titration of DRT to produce the desired improvement in motor symptoms while minimizing the associated detrimental effects. This is important in order to maximize the motor re-learning based on repetition, reward and practice during rehabilitation. In this scenario, we review the knowledge concerning the interactions between DRT, learning and reward, examine the most impactful DRT side effects and provide suggestions for optimizing rehabilitation in PD.
Open Access
An exploratory study on the value of service learning projects and their impact on community service involvement and critical thinking
(2007) Joseph, M.; Stone, G. W.; Grantham, K.; Harmancioglu, N.; Ibrahim, E.
Purpose - This exploratory study attempts to capture some of the principal benefits/factors attributable to service learning/community service projects, from a student perspective. Design/methodology/approach - A sample of 67 males and 83 females (16 graduate, 71 seniors, and 63 juniors) participated in the study. Findings - Students believe that their college experience is preparing them for the job market, that critical thinking has been enhanced, and that their college academic experience has emphasized community service upon graduation. Practical implications - The results increase one's knowledge of the benefits of service learning since so much emphasis is currently being placed on improving the critical thinking and problem-solving ability of undergraduate business students. Originality/value - Practitioners would be interested in understanding the impact that service learning can have on the problem-solving ability of potential employees. If additional research could advance the proposition that students with service learning experience are generally superior in terms of their problem-solving skills to students with no similar experience, then evidence of a service learning component on a student résumeé suddenly adds value to the employer.
Open Access
From hate to love: how learning can change affective responses to touched materials
(Springer Science and Business Media Deutschland GmbH, 2020) Cavdan, M.; Freund, A.; Trieschmann, A.-K.; Doerschner, Katja; Drewing, K.; Nisky, I.; Hartcher-O’Brien, J.; Wiertlewski, M.; Smeets, J.
People display systematic affective reactions to specific properties of touched materials. For example, granular materials such as fine sand feel pleasant, while rough materials feel unpleasant. We wondered how far such relationships between sensory material properties and affective responses can be changed by learning. Manipulations in the present experiment aimed at unlearning the previously observed negative relationship between roughness and valence and the positive one between granularity and valence. In the learning phase, participants haptically explored materials that are either very rough or very fine-grained while they simultaneously watched positive or negative stimuli, respectively, from the International Affective Picture System (IAPS). A control group did not interact with granular or rough materials during the learning phase. In the experimental phase, participants rated a representative diverse set of 28 materials according to twelve affective adjectives. We found a significantly weaker relationship between granularity and valence in the experimental group compared to the control group, whereas roughness-valence correlations did not differ between groups. That is, the valence of granular materials was unlearned (i.e., to modify the existing valence of granular materials) but not that of rough materials. These points to differences in the strength of perceptuo-affective relations, which we discuss in terms of hard-wired versus learned connections.
Open Access
Jamming bandits-a novel learning method for optimal jamming
(Institute of Electrical and Electronics Engineers Inc., 2016) Amuru, S.; Tekin, C.; Van Der Schaar, M.; Buehrer, R.M.
Can an intelligent jammer learn and adapt to unknown environments in an electronic warfare-type scenario? In this paper, we answer this question in the positive, by developing a cognitive jammer that adaptively and optimally disrupts the communication between a victim transmitter-receiver pair. We formalize the problem using a multiarmed bandit framework where the jammer can choose various physical layer parameters such as the signaling scheme, power level and the on-off/pulsing duration in an attempt to obtain power efficient jamming strategies. We first present online learning algorithms to maximize the jamming efficacy against static transmitter-receiver pairs and prove that these algorithms converge to the optimal (in terms of the error rate inflicted at the victim and the energy used) jamming strategy. Even more importantly, we prove that the rate of convergence to the optimal jamming strategy is sublinear, i.e., the learning is fast in comparison to existing reinforcement learning algorithms, which is particularly important in dynamically changing wireless environments. Also, we characterize the performance of the proposed bandit-based learning algorithm against multiple static and adaptive transmitter-receiver pairs.
Open Access
Learning by imitation
(Elsevier BV, 1999) Başçı, E.
This paper introduces a learning algorithm that allows for imitation in recursive dynamic games. The Kiyotaki-Wright model of money is a well-known example of such decision environments. In this context, learning by experience has been studied before. Here, we introduce imitation as an additional channel for learning. In numerical simulations, we observe that the presence of imitation either speeds up social convergence to the theoretical Markov-Nash equilibrium or leads every agent of the same type to the same mode of suboptimal behavior. We observe an increase in the probability of convergence to equilibrium, as the incentives for optimal play become more pronounced.
Open Access
Learning the optimum as a Nash equilibrium
(Elsevier BV, 2000) Özyıldırım, S.; Alemdar, N. M.
This paper shows the computational benefits of a game theoretic approach to optimization of high dimensional control problems. A dynamic noncooperative game framework is adopted to partition the control space and to search the optimum as the equilibrium of a k-person dynamic game played by k-parallel genetic algorithms. When there are multiple inputs, we delegate control authority over a set of control variables exclusively to one player so that k artificially intelligent players explore and communicate to learn the global optimum as the Nash equilibrium. In the case of a single input, each player's decision authority becomes active on exclusive sets of dates so that k GAs construct the optimal control trajectory as the equilibrium of evolving best-to-date responses. Sample problems are provided to demonstrate the gains in computational speed and accuracy. © 2000 Elsevier Science B.V.
Restricted
Learning without knowing
(1988) Shklar, Judith N.
Open Access
Learning, inflation, and the Phillips Curve
(2000) Kaplan, Duygu
The Replicator Dynamics of Evolutionary Game Theory is introduced in a closed economy so as to model how a continuum of firms evolve over time with respect to the pricing strategies. Incorporation- of Replicator Dynamics facilitates modelling microeconomic frictions that lead to a Phillips Curve on the macroeconomic level. I'he firms are boundedly rational players which are learning, and are apt to make mistakes. Mistakes function as a mutation process and prevent a strategy from becoming extinct. An arbitrary non-empty set of consumers face a cash-in-advance constraint and total consumption spending is symmetrically affected by changes in growth rate of money supply which is stochastic. Using a discrete price set, we introduce heterogeneity of firm behaviour in a single homogenous good market. The economy is simulated ibr a large number of finitely many time periods. A Phillips Curve type linkage between infiation and output is recognized at stationary states. The slope of the Phillips Curve is observed to increase as mean of money growth rate gets higher or as the uncertainty in money growth rate is increased. Slope of the Phillips Curve diminishes as price stickiness is intensified by either reducing the mistake level or by increasing the firms' responsiveness to relative payoff realizations.
Open Access
A model of boundedly rational learning in dynamic games
(1997) Aksoy, Hakan
There are various computer-based algorithms about boundedly rational players’ learning how to behave in dynamic games, including classifier systems, genetic algorithms and neural networks. Some examples of studies using boundedly rational players are Axelrod (1987), Miller (1989), Andreoni and Miller (1990) who use genetic algorithm and Marimon etal. (1990) and Arthur (1990) who use classifier systems. In this dissertation, a Two Armed Bandit Problem and the KiyotakiWright (1989) Economic Environment are constructed and the learning behaviour ol the boundedly rational players is observed by using classifier systems in computer programs. From the simulation results, we observe that experimentation and imitation enables faster convergence to the correct decision rules of players in both repeated static decision problems and dynamic games.
Open Access
Neural network-based target differentiation using sonar for robotics applications
(IEEE, 2000-08) Barshan, B.; Ayrulu, B.; Utete, S. W.
This study investigates the processing of sonar signals using neural networks for robust differentiation of commonly encountered features in indoor robot environments. The neural network can differentiate more targets with higher accuracy, improving on previously reported methods. It achieves this by exploiting the identifying features in the differential amplitude and time-of-flight (TOF) characteristics of these targets. Robustness tests indicate that the amplitude information is more crucial than TOF for reliable operation. The study suggests wider use of neural networks and amplitude information in sonar-based mobile robotics.
Open Access
Neural networks for improved target differentiation and localization with sonar
(Pergamon Press, 2001) Ayrulu, B.; Barshan, B.
This study investigates the processing of sonar signals using neural networks for robust differentiation of commonly encountered features in indoor robot environments. Differentiation of such features is of interest for intelligent systems in a variety of applications. Different representations of amplitude and time-of-flight measurement patterns acquired from a real sonar system are processed. In most cases, best results are obtained with the low-frequency component of the discrete wavelet transform of these patterns. Modular and non-modular neural network structures trained with the back-propagation and generating-shrinking algorithms are used to incorporate learning in the identification of parameter relations for target primitives. Networks trained with the generating-shrinking algorithm demonstrate better generalization and interpolation capability and faster convergence rate. Neural networks can differentiate more targets employing only a single sensor node, with a higher correct differentiation percentage (99%) than achieved with previously reported methods (61-90%) employing multiple sensor nodes. A sensor node is a pair of transducers with fixed separation, that can rotate and scan the target to collect data. Had the number of sensing nodes been reduced in the other methods, their performance would have been even worse. The success of the neural network approach shows that the sonar signals do contain sufficient information to differentiate all target types, but the previously reported methods are unable to resolve this identifying information. This work can find application in areas where recognition of patterns hidden in sonar signals is required. Some examples are system control based on acoustic signal detection and identification, map building, navigation, obstacle avoidance, and target-tracking applications for mobile robots and other intelligent systems. Copyright © 2001 Elsevier Science Ltd.
Open Access
Novel object recognition is not affected by age despite age-related brain changes
(Scientific Research Publishing, 2013) Aktoprak, İlay; Dinç, Pelin; Günay, Gizem; Adams, Michelle M.
Age-related memory impairments show a progressive decline across lifespan. Studies have demonstrated equivocal results in biological and behavioral outcomes of aging. Thus, in the present study we examined the novel object recognition task at a delay period that has been shown to be impaired in aged rats of two different strains. Moreover, we used a strain of rats, Fisher 344XBrown Norway, which have published age-related biological changes in the brain. Young (10 month old) and aged (28 month old) rats were tested on a standard novel object recognition task with a 50-minute delay period. The data showed that young and aged rats in the strain we used performed equally well on the novel object recognition task and that both young and old rats demonstrated a righthanded side preference for the novel object. Our data suggested that novel object recognition is not impaired in aged rats although both young and old rats have a demonstrated side preference. Thus, it may be that genetic differences across strains contribute to the equivocal results in behavior, and genetic variance likely influences the course of cognitive aging.