Native palms, such as the Sabal palmetto, play an important role in maintaining the ecological balance in Florida. As a side-effect of modern globalization, new phytopathogens like Texas Phoenix Palm Decline have been introduced into forest systems that threaten native palms. This presents new challenges for forestry managers and geographers. Advances in remote sensing has assisted the practice of forestry by providing spatial metrics regarding the type, quantity, location, and the state of heath for trees for many years. This study provides spatial details regarding the general palm decline in Florida by taking advantage of the new developments in deep learning constructs coupled with high resolution WorldView-2 multispectral/temporal satellite imagery and LiDAR point cloud data. A novel approach using TensorFlow deep learning classification, multiband spatial statistics and indices, data reduction, and step-wise refinement masking yielded a significant improvement over Random Forest classification in a comparison analysis. The results from the TensorFlow deep learning were then used to develop an Empirical Bayesian Kriging continuous raster as an informative map regarding palm decline zones using Normalized Difference Vegetation Index Change. The significance from this research showed a large portion of the study area exhibiting palm decline and provides a new methodology for deploying TensorFlow learning for multispectral satellite imagery.

Functional magnetic resonance imaging (fMRI) technology has been used in the investigation of human brain functionality and assist in brain disease diagnosis. While fMRI can be used to model both spatial and temporal brain functionality, the analysis of the fMRI images and the discovery of patterns for certain brain diseases is still a challenging task in medical imaging. Deep learning has been used more and more in medical field in an effort to further improve disease diagnosis due to its effectiveness in discovering high-level features in images. Convolutional neural networks (CNNs) is a class of deep learning algorithm that have been successfully used in medical imaging and extract spatial hierarchical features. The application of CNNs in fMRI and the extraction of brain functional patterns is an open field for research. This project focuses on how fMRIs can be used to improve Autism Spectrum Disorders (ASD) detection and diagnosis with 3D resting-state functional MRI (rs-fMRI) images. ASDs are a range of neurodevelopment brain diseases that mostly affect social function. Some of the symptoms include social and communicating difficulties, and also restricted  and repetitive  behaviors. The  symptoms appear on early childhood and tend to develop in time thus an early diagnosis is required. Finding a proper model for identifying between ASD and healthy subject is a challenging task and involves a lot of hyper-parameter tuning. In this project a grid search approach is followed in the quest of the optimal CNN architecture. Additionally, regularization and augmentation techniques are implemented in an effort to further improve the models performance.

When asked to implement a neural network application, the decision concerning what hardware platform to use may not always be easily made. This thesis studies various relevant platforms with regards to performance, power efficiency and usability, with the purpose of providing a basis for such decisions. The hardware platforms which are studied were a GPU, an FPGA and a CPU. The project implements Convolutional Neural Networks (CNN) on the different hardware platforms using several tools and frameworks. The final implementation uses BNN-PYNQ for the implementation on the FPGA and CPU, which provided ready-to-run code and overlays for quantized CNNs and fully connected neural networks. Next, these networks are copied using TensorFlow, and optimized to FP32, FP16 and INT8 precision using TensorRT for use on the GPU. The results indicate that the FPGA outperforms the GPU with a factor of 100 for the CNN networks, and a factor of 1000 on the fully connected networks with regards to inference speed. For power efficiency, the FPGA again outperforms the GPU. The thesis concludes that for a neural network application, an FPGA is preferred if performance is a priority. However, the GPU proved to have a greater ease of use due to the many tools and frameworks available. If easy implementation and high design flexibility is a priority, a GPU is instead recommended.

Intraday stock trading is an infamously difficult and risky strategy. Momentum and reversal strategies and long short-term memory (LSTM) neural networks have been shown to be effective for selecting stocks to buy and sell over time periods of multiple days. To explore whether these strategies can be effective for intraday trading, their implementations were simulated using intraday price data for stocks in the S&P 500 index, collected at 1-second intervals between February 11, 2021 and March 9, 2021 inclusive. The study tested 160 variations of momentum and reversal strategies for profitability in long, short, and market-neutral portfolios, totaling 480 portfolios. Long and short portfolios for each strategy were also compared to the market to observe excess returns. Eight reversal portfolios yielded statistically significant profits, and 16 yielded significant excess returns. Tests of these strategies on another set of 16 days failed to yield statistically significant returns, though average returns remained profitable. Four LSTM network configurations were tested on the same original set of days, with no strategy yielding statistically significant returns. Close examination of the stocks chosen by LSTM networks suggests that the networks expect stocks to exhibit a momentum effect. Further studies may explore whether an intraday reversal effect can be observed over time during different market conditions and whether different configurations of LSTM networks can generate significant returns.

No description available.

Multi-robot networks are used in a variety of military and civilian applications such as harbour protection, perimeter surveillance, search & rescue missions and cooperative estimation, among others. In order to develop functional multi-robot networks to achieve these tasks, a combination of theoretical and experimental work is required. Theoretical research aims to model the open and closed loop dynamics of multi-robot systems and to develop corresponding control algorithms for tackling the previously mentioned tasks. Experimental work focuses on the hardware or simulated implementations to test and evaluate the algorithms’ performance, and eventually inform refinements of theoretical algorithms to adapt to hardware imposed intrinsic constraints. As theoretical and algorithmic research in the field of multi-robot networks matures, a need for experimental validation of a variety of sophisticated algorithmic tools becomes apparent, with the two aspects co-developing to inform theoretical refinements that account for hardware induced constraints, and possible technological advances suggested by theoretical predictions. This thesis contributes a design for a hardware test platform for evaluating and implementing algorithms in the field of multi-robot networks. The test platform design implements an off the market robot solution for the robotic agents, discussing various additional embedded frameworks that allow for capabilities such as indoor agent localization, inter-agent wireless communication and agent locomotion, with the goal of understanding if a combination of existing market and academic technologies can be used to develop a cost effective hardware multi-agent test platform. The proposed hardware design is then validated on previously developed multiagent area coverage control and optimization algorithms. More specifically, the hardware test platform is used to implement various optimal coverage problems with time invariant risk density distributions. Both uniform and nonuniform risk density distribution scenarios are considered. These experimental results are compared with simulations to assess if the proposed hardware test platform design can plausibly reproduce behaviours that are consistent with theoretical predictions of area coverage control algorithms. Future work will include the extension of the testing capabilities of this test platform to a larger class of multi-agent control algorithms.

A current field of research today is machine learning because it can simplify everyday life for human beings. A functioning system that has learned specific tasks can make it easier for companies in both cost and time. A company who want to use machine learning is SCA, who owns and manages forests to produce products. They have a need to automate forest classification. In order to evaluate forests, and to plan forestry measures, the proportion of leafy tree that is not used in production must be determined. Today, manual work is required of people who have to investigate aerial photos to classify the tree types. This study investigates whether it is possible, through machine learning, to teach a computer to determine whether it is leaf or not in photographs. A program is constructed with the library ImageAI which receives methods for training and predicting information in images. It examines how the choice of neural network and the number of images affects the safety of the models and how reliable the models can be. Exercise time and hardware are also two factors that are investigated. The result shows that the neural network ResNet delivers the safest results and the more images the computer exercises, the safer the result. The final model is a ResNet model that has trained on 20,000 images and has 79,0 percent security. Based on 50 samples, the mean value for safety is 90,5 percent and the median is 99,6 percent. / Maskininlärning är idag ett aktuellt forskningsområde som kan förenkla vardagen för oss människor. Ett fungerande system som har lärt sig specifika uppgifter kan underlätta för företag i både kostnad och tid. Ett företag som vill använda maskininlärning är SCA, som äger och förvaltar skog för att producera produkter. De har behov av att automatisera klassificering av skog. För att värdera skogar, samt planera skogsåtgärder, måste andelen lövträd som inte används i produktionen bestämmas. Idag krävs det manuellt arbete av personer som måste undersöka flygfoton för att klassificera trädtyperna. Denna studie undersöker om det är möjligt, via maskininlärning, att lära en dator avgöra om det är löv eller inte i ortofoton. Ett program konstrueras med biblioteket ImageAI som erhåller metoder för att träna och förutsäga information i bilder. Det undersöks hur valet av neuralt nätverk och antalet bilder påverkar säkerheten för modellerna samt hur tillförlitlig modellerna kan bli. Träningstid och hårdvara är också två faktorer som studeras. Resultatet visar att neurala nätverket ResNet levererar säkrast resultat och desto fler bilder datorn tränar på, desto säkrare blir resultatet. Den slutgiltiga modellen är en ResNet-modell som tränat på 20 000 bilder och har 79,0 procents säkerhet. Utifrån 50 stickprov är medelvärdet för säkerheten 90,5 procent och medianen 99,6 procent.

本研究提出了一個類神經網路機制，可以及時有效的預測利差交易(carry trade)的收益。為了實現及時性，我們將通過Tensorflow和圖形處理單元(GPU)來實作這個機制。此外，類神經網路機制需要處理具有概念飄移和異常值的時間序列數據。而我們將透過設計的實驗來驗證這個機制的及時性與有效性。 在實驗過程中，我們發現在演算法設置不同的參數將影響類神經網路的性能。本研究將討論不同參數下所產生的不同結果。實驗結果表明，我們所提出的類神經網路機制可以預測出利差交易的收益的動向。希望這個研究將對機器學習和金融領域皆有所貢獻。 / This research derives an artificial neural networks (ANN) mechanism for timely and effectively predicting the return of carry trade. To achieve the timeliness, the ANN mechanism is implemented via the infrastructure of TensorFlow and graphic processing unit (GPU). Furthermore, the ANN mechanism needs to cope with the time series data that may have concept-drifting phenomenon and outliers. An experiment is also designed to verify the timeliness and effectiveness of the proposed mechanism. During the experiment, we find that different parameters we set in the algorithm will affect the performance of the neural network. And this research will discuss the different results in different parameters. Our experiment result represents that the proposed ANN mechanism can predict movement of the returns of carry trade well. Hope this research would contribute for both machine learning and finance field.

本研究要探討的是在新聞、論壇和社群媒體討論的相關訊息是否真的會影響匯率的運動的假設。對於這樣的研究目標，我們建立了一個實驗，首先以文字探勘技術應用在新聞、論壇與社群媒體來產生與匯率相關的數值表示。接著，機器學習技術應用於學習得到的數值表示和匯率波動之間的關係。最後，我們證明透過檢驗所獲得的關係的有效性的假設。在此研究中，我們提出一種兩階段的神經網路來學習與預測每日美金兌台幣匯率的走勢。不同於其他專注於新聞或者社群媒體的研究，我們將他們進行整合，並將論壇的討論納為輸入資料。不同的資料組合產生出多種觀點，而三個資料來源的不同組合可能會以不同的方式影響預測準確率。透過該方法，初步實驗的結果顯示此方法優於隨機漫步模型。 / This study wants to explore the hypothesis that the relevant information in the news, the posts in forums and discussions on the social media can really affect the daily movement of exchange rates. For such study objective, we set up an experiment, where the text mining technique is first applied to the news, the forum and the social media to generate numerical representations regarding the textual information relevant with the exchange rate. Then the machine learning technique is applied to learn the relationship between the derived numerical representations and the movement of exchange rates. At the end, we justify the hypothesis through examining the effectiveness of the obtained relationship. In this paper, we propose a hybrid neural networks to learn and forecast the daily movements of USD/TWD exchange rates. Different from other studies, which focus on news or social media, we integrate them and add the discussion of forum as input data. Different data combinations yield many views while different combination of three data sources might affect the forecasting accuracy rate in different ways. As a result of this method, the experiment result was better than random walk model.

This thesis deals with the automatic detection of elevator controls in personal elevators through digital imaging using computer vision. The theoretical part of the thesis goes through methods of image processing with regards to object detection in image and research of previous solutions. This leads to investigation into the field of convolutional neural networks. The practical part covers the creation of elevator controls image dataset, selection, training and evaluation of the used models and the implementation of a robust algorithm utilizing the detection of elevator controls. The conclussion of the work discusses the suitability of the detection on given task.