The mammalian visual system is an incredibly complex computation device, capable of performing the various tasks of seeing: navigation, pattern and object recognition, motor coordination, trajectory extrapolation, among others. Decades of research has shown that experience-dependent plasticity of cortical circuitry underlies the impressive ability to rapidly learn many of these tasks and to adjust as required. One particular thread of investigation has focused on unsupervised learning, wherein changes to the visual environment lead to corresponding changes in cortical circuits. The most prominent example of unsupervised learning is ocular dominance plasticity, caused by visual deprivation to one eye and leading to a dramatic re-wiring of cortex. Other examples tend to make more subtle changes to the visual environment through passive exposure to novel visual stimuli. Here, we use one such unsupervised paradigm, sequence learning, to study experience-dependent plasticity in the mouse visual system. Through a combination of theory and experiment, we argue that the mammalian visual system is an unsupervised learning device. Beginning with a mathematical exploration of unsupervised learning in biology, engineering, and machine learning, we seek a more precise expression of our fundamental hypothesis. We draw connections between information theory, efficient coding, and common unsupervised learning algorithms such as Hebbian plasticity and principal component analysis. Efficient coding suggests a simple rule for transmitting information in the nervous system: use more spikes to encode unexpected information, and fewer spikes to encode expected information. Therefore, expectation violations ought to produce prediction errors, or brief periods of heightened firing when an unexpected event occurs. Meanwhile, modern unsupervised learning algorithms show how such expectations can be learned. Next, we review data from decades of visual neuroscience research, highlighting the computational principles and synaptic plasticity processes that support biological learning and seeing. By tracking the flow of visual information from the retina to thalamus and primary visual cortex, we discuss how the principle of efficient coding is evident in neural activity. One common example is predictive coding in the retina, where ganglion cells with canonical center-surround receptive fields compute a prediction error, sending spikes to the central nervous system only in response to locally-unpredictable visual stimuli. This behavior can be learned through simple Hebbian plasticity mechanisms. Similar models explain much of the activity of neurons in primary visual cortex, but we also discuss ways in which the theory fails to capture the rich biological complexity. Finally, we present novel experimental results from physiological investigations of the mouse primary visual cortex. We trained mice by passively exposing them to complex spatiotemporal patterns of light: rapidly-flashed sequences of images. We find evidence that visual cortex learns these sequences in a manner consistent with efficient coding, such that unexpected stimuli tend to elicit more firing than expected ones. Overall, we observe dramatic changes in evoked neural activity across days of passive exposure. Neural responses to the first, unexpected sequence element increase with days of training while responses at other, expected time points either decrease or stay the same. Furthermore, substituting an unexpected element for an expected one or omitting an expected element both cause brief bursts of increased firing. Our results therefore provide evidence for unsupervised learning and efficient coding in the mouse visual system, especially because unexpected events drive prediction errors. Overall, our analysis suggests novel experiments, which could be performed in the near future, and provides a useful framework to understand visual perception and learning.

<p> Multiple-subcarrier modulated (MSM) indoor wireless optical systems provide highspeed bandwidth-efficient communication as well as multi-access capability at low cost. A major drawback of such systems is the high average optical power requirement. To address this problem, trellis codes are designed for data transmission realizing an average optical power reduction as high as 0.95 dB. A simultaneous peak optical power reduction up to 0.44 dB is also possible by employing this technique.</p> <p> Moreover, the unregulated bandwidth available in wireless optical channels is exploited to reduce the average optical power in MSM wireless optical systems. Data transmission is confined in a bandwidth located near DC, while reserved carrier signals are designed outside the data bandwidth to reduce the average optical power. Distorted out-of-band signals at higher frequency are removed at the receiver by low-pass filtering and not used for detection. To design the out-of-band carrier amplitudes, both a standard optimization algorithm over the set of real numbers and an exhaustive search over discrete constellation are proposed. It is shown that significant average optical power reduction as high as 2.56 dB is achieved.</p> <p> It is assumed that a symbol-by-symbol bias is used in MSM wireless optical systems since it offers significant optical power reduction. The received DC bias level provides a degree of diversity and is used at the receiver to improve detection. It is also used to design trellis codes with better distance properties. In this manner, an additional average optical power reduction up to 0.50 dB is achieved with a simultaneous peak power reduction of 0.46 dB for MSM wireless optical systems.</p> / Thesis / Master of Applied Science (MASc)

Определение разумного масштаба деятельности предприятия является необходимым условием его эффективного функционирования. Цель магистерской диссертации является совершенствование методического инструментария обоснования решений по изменению масштаба предприятия для повышения экономической эффективности его деятельности. В работе рассматривается генезис понятия предельных издержек и вопросы взаимосвязи масштаба предприятия и эффективности его деятельности. В качестве источников информации для апробации методического подхода использовались годовые отчёты предприятий. В магистерской диссертации был разработан методический подход к обоснованию решения о целесообразности наращивания масштаба предприятия и выбору конкретных направлений увеличения масштаба, основанный на анализе динамики вовлечения производственных факторов в производство и оценке устойчивых и существенных изменений их отдачи, вызванных ростом предельных издержек в долгосрочном периоде. / Determination of a reasonable scale of the enterprise is a necessary condition for its effective functioning. The purpose of the master's thesis is to improve the methodological support for substantiating decisions on changing the scale of an enterprise, which will increase the economic efficiency of its activities. The work examines the genesis of the concept of marginal costs and the relationship between the scale of the enterprise and the efficiency of its activities. In the master's thesis, a methodological approach was developed to substantiate the decision on the feasibility of increasing the scale of the enterprise and the choice of specific directions for increasing the scale. It is based on the analysis of the dynamics of the involvement of production factors in production and the assessment of stable and significant changes in their returns, caused by an increase in marginal costs in the long run. Annual reports of enterprises were used as sources of information for testing the methodological approach.

No description available.

Introduction– The purpose of the study was to be able to examine and present energy-efficient solutions for future new constructions for multi-dwelling buildings with aim on design. It was based on investigating the causes and reasons for the high energy consumption in apartment buildings, as well as investigating the connection between energy change and environmental impact. Method – The chosen method for this work was a case study which is a qualitative research method. The method had a purpose that focuses on identifying problems and describing in depth how, why, and which measures are suitable for reducing energy consumption in apartment buildings. The work combined quantitative and qualitative research methods. For a better understanding of the research topic, a qualitative literature review of various literature and scientific reports was carried out. There, a certain number of interviews were conducted with several architects and engineers where questions about the research topic were asked. Results – Under the results chapter, the standards, and requirements as well as other building specifications for multi-dwelling buildings that affected energy were presented. Furthermore, several solution possibilities and techniques were presented. In addition, the interview answers that have the clearest connection to the question. Finally, it led to the primary energy number not being allowed to rise above 75 [kWh/m2 Atemp and year], a good facade material choice can be brick and concrete. Certain technical solutions can contribute to less energy consumption such as air heat exchange and solar cells, and the environmental impact is connected with energy consumption. Analysis – During the analysis of empirical evidence, it is pointed out the factors that cause high energy consumption in apartment buildings. In addition, the solution options that led to a reduction in energy consumption and energy efficiency for apartment buildings are revealed. Furthermore, the connection between change in energy consumption and environmental impact is brought together. Discussion – The case study showed that there were several factors behind high energy consumption in apartment buildings, partly linked to design, partly linked to socio-economic conditions. Based on both literatures, the study and interviews, reliable data could be obtained to answer questions. Furthermore, the chapter contains advantages and disadvantages of using the case study method and data collection techniques. Keywords: Energy reduction, Energy consumption, Energy efficiency, Apartment building, Solution possibility, Environmental impact, new constructions, Design. / Introduktion – Syftet med studien var att kunna undersöka och presentera fram energieffektiva lösningar för framtida nybyggnationer för flerbostadshus med fokus på utformning. Det var med utgångpunkt på att undersöka orsaker och anledningar till de höga energiförbrukningarna i flerbostadshus, samt att undersöka sambandet mellan energiförändring och miljöpåverkan Metod – Den valda metoden för detta arbete var en fallstudie som är en kvalitativ forskningsmetod. Metoden har ett syfte som fokuserar på att identifiera problem och beskriva djupgående kring hur, varför och vilka åtgärder som är lämpliga för minskning av energiförbrukning i flerbostadshus. I arbetet kombinerades kvantitativa och kvalitativa undersökningsmetoder. För en bredare förståelses kring forskningsämnet så utfördes en kvalitativ litteraturstudie av olika litteraturer och vetenskapliga rapporter. Därpå genomfördes intervjuer med arkitekter och ingenjörer där frågor kring forskningsämnet ställdes. Resultat – Under resultatkapitel presenteras de standarder och krav samt andra byggspecifikationer för flerbostadshus som berörde energi. Vidare uppvisades det flertal lösningsmöjligheter och tekniker. Dessutom de intervjusvar som har tydligast koppling till frågeställningarna. Slutligen lede det till att det primära energitalet få inte stiga över 75 [kWh/m2 Atemp och år], ett bra fasad materialval kan vara tegel och betong, visa tekniska lösningar kan bidra till mindre energiförbrukning liksom luftvärmeväxling och solceller, samt miljöpåverkan hänger ihop med energiförbrukningen.      Analys – Under analys av empiri så påpekas det på de faktorer som orsakar hög energiförbrukning inom flerbostadshus. Dessutom så tyder de lösningsmöjligheter som leder till minskning av energiförbrukning och effektivisering av energi för flerbostadshus. Vidare i kapitlet sammanfördes sambanden mellan förändring i energiförbrukning och miljöpåverkan. Diskussion – Fallstudien visade att det finns flera faktorer bakom hög energiförbrukning i flerbostadshus, dels det som var kopplat till utformning, dels kopplat till socioekonomiska förhållanden. Utifrån både litteraturstudie och intervjuerna kunde trovärdiga data inhämtas för att besvara frågeställningarna. Vidare så innehåller kapitlet fördelar och nackdelar med användning av fallstudie som metod och datainsamlingsteknikerna. Nyckelord – Energireduktion, Energiförbrukning, Energieffektivitet, Flerbostadshus, Lösningsmöjlighet, Miljöpåverkan, Nybyggnationer, Utformning.

The ability to understand videos at the level of pixels plays a key role in a wide range of computer vision applications. For example, a robot or autonomous vehicle relies on classifying each pixel in the video stream into semantic categories to holistically understand the surrounding environment, and video editing software needs to exploit the spatiotemporal context of video pixels to generate various visual effects. Despite the great progress of Deep Learning (DL) techniques, applying DL-based vision models to process video pixels remains practically challenging, due to the high volume of video data and the compute-intensive design of DL approaches. In this thesis, we aim to design efficient and robust deep models for pixel-level video understanding of high-level semantics, mid-level grouping, and low-level interpolation. Toward this goal, in Part I, we address the semantic analysis of video pixels with the task of Video Semantic Segmentation (VSS), which aims to assign pixel-level semantic labels to video frames. We introduce methods that utilize temporal redundancy and context to efficiently recognize video pixels without sacrificing performance. Extensive experiments on various datasets demonstrate our methods' effectiveness and efficiency on both common GPUs and edge devices. Then, in Part II, we show that pixel-level motion patterns help to differentiate video objects from their background. In particular, we propose a fast and efficient contour-based algorithm to group and separate motion patterns for video objects. Furthermore, we present learning-based models to solve the tracking of objects across frames. We show that by explicitly separating the object segmentation and object tracking problems, our framework achieves efficiency during both training and inference. Finally, in Part III, we study the temporal interpolation of pixels given their spatial-temporal context. We show that intermediate video frames can be inferred via interpolation in a very efficient way, by introducing the many-to-many splatting framework that can quickly warp and fuse pixels at any number of arbitrary intermediate time steps. We also propose a dynamic refinement mechanism to further improve the interpolation quality by reducing redundant computation. Evaluation on various types of datasets shows that our method can interpolate videos with state-of-the-art quality and efficiency. To summarize, we discuss and propose efficient pipelines for pixel-level video understanding tasks across high-level semantics, mid-level grouping, and low-level interpolation. The proposed models can contribute to tackling a wide range of real-world video perception and understanding problems in future research.

With the increase of global temperature and decrease of fossil fuel sources, biofuels become an excellent alternative in present days. Because of its oxygenated nature, biofuels are found to be more environmentally friendly over fossil fuels. Therefore, in this study, initially two different biofuels: ethanol and 2,5 dimethyl furan (DMF) are considered as an additive to gasoline which shows a significant improvement in its combustion characteristics. Due to this promising result for further studies of these biofuel, details chemical kinetic study of biofuels is considered in this work through generating its mechanism for engine relevant conditions. Detail chemical mechanism PCRL-Mech1 is generated for ethanol which is applicable for wide range of operating conditions. The mechanism is successfully validated with available experimental data of laminar burning speed (LBS) and ignition delay time (IDT). Species concentration at different reactor conditions are also considered for the comparison which shows an excellent agreement. Detail mechanism generation for another newer biofuel anisole is also considered because of its favorable features in combustion properties and potential source of biomass. Anisole is a higher hydrocarbon aromatic component and comparative newer fuel which has limited experimental data. However, with that available experimental data, the developed anisole mechanism shows a good agreement predicting LBS and IDT results. The chemical kinetics of this fuel is also analyzed through reaction path flux and sensitivity analyses. Although, detail mechanisms have higher accuracy, they would be very expensive when using in multiscale computational fluid dynamics (CFD) modeling. Therefore, different mechanism reduction schemes are considered to reduce the mechanism size. Initially direct relation graph (DRG), direct relation graph with error propagation (DRGEP) and sensitivity analysis is implemented to generate a skeleton mechanism for PCRL-Mech1, which successfully reduced its size. In addition, the rate-controlled constraint equilibrium (RCCE) analysis is considered as a reduction scheme. The constraints for RCCE calculation are selected through approximate singular value decomposition of actual degree of disequilibrium (ASVDADD) analysis. A good comparison of temperature profile of RCCE simulation proves the success of ASVDADD method.

The State of Mississippi wants to manage its oyster resource to increase production, quality, ecological, and economic benefits. In this study, we employ modern portfolio theory (MPT) to test if there are potential gains to hold multiple oyster resources for multiple benefits to aid the state's effort in achieving its goal. Using a Delphi approach, we elicit complete sets of data on ecosystem services (on oxygen, nutrients, sedimentation, and salinity) across multiple oyster resources (traditional plantings, off-bottom farms, and restored reefs). A benefit transfer method is used later to assigned money-metric value to each service estimate. The multiple service values are then aggregated into net service value. We compute the means, standard deviations, and correlations of benefits across all resources using the net service values, and generate efficient frontiers from that information. Results indicate that Mississippi could benefit from holding multiple oyster resources while focusing more on off-bottom oyster farms.

No description available.

No description available.