With the rise of inverter-based resources (IBRs) within the power system, the control of grid-connected converters (GCC) has become pertinent due to the fact they interface IBRs to the grid. The conventional method of control for grid-connected converters (GCCs) such as the voltage-sourced converter (VSC) is through a decoupled control loop in the synchronous reference frame. However, this model-based control method is sensitive to parameter changes causing deterioration in controller performance. Data-driven approaches such as machine learning can be utilized to design controllers that are capable of operating GCCs in various system conditions. This work reviews different machine learning applications in power systems as well as the conventional method of controlling a VSC. It explores a deep learning-based control method for a three-phase grid-connected VSC, specifically utilizing a long short-term memory (LSTM) network for robust control. Simulations of a conventional controlled VSC are conducted using Simulink to collect data for training the LSTM-based controller. The LSTM model is built and trained using the Keras and TensorFlow libraries in Python and tested in Simulink. The performance of the LSTM-based controller is evaluated under different case studies and compared to the conventional method of control. Simulation results demonstrate the effectiveness of this approach by outperforming the conventional controller and maintaining stability under different system parameter changes. / Master of Science / The desire to minimize the use of fossil fuels and reduce carbon footprints has increased the usage of renewable energies also known as inverter-based resources (IBRs) within the power grid. These resources add a level of complexity to operating the grid due to the fluctuating nature of IBRs and are connected to the power grid through grid-connected converters (GCCs). The control method conventionally used for GCCs is derived by accounting for the system parameters, creating a mathematical model under constant parameters. However, the parameters of the system are susceptible to changes under different operating and environmental conditions. This results in poor performance from the controller under various operating conditions due to its inability to be adaptive to the system. Data-driven approaches such as machine learning are becoming increasingly popular for their ability to capture the dynamics of a system with limited knowledge. The different applications of machine learning within power systems include fault diagnosis, energy management, and cyber security. This work explores the use of utilizing deep learning techniques for a robust approach of controlling GCCs.

Load models have evolved from simple ZIP model to composite model that incorporates the transient dynamics of motor loads. This research utilizes the latest trend on Machine Learning and builds reliable and accurate composite load model. A composite load model is a combination of static (ZIP) model paralleled with a dynamic model. The dynamic model, recommended by Western Electricity Coordinating Council (WECC), is an induction motor representation. In this research, a dual cage induction motor with 20 parameters pertaining to its dynamic behavior, starting behavior, and per unit calculations is used as a dynamic model. For machine learning algorithms, a large amount of data is required. The required PMU field data and the corresponding system models are considered Critical Energy Infrastructure Information (CEII) and its access is limited. The next best option for the required amount of data is from a simulating environment like PSSE. The IEEE 118 bus system is used as a test setup in PSSE and dynamic simulations generate the required data samples. Each of the samples contains data on Bus Voltage, Bus Current, and Bus Frequency with corresponding induction motor parameters as target variables. It was determined that the Artificial Neural Network (ANN) with multivariate input to single parameter output approach worked best. Recurrent Neural Network (RNN) is also experimented side by side to see if an additional set of information of timestamps would help the model prediction. Moreover, a different definition of a dynamic model with a transfer function-based load is also studied. Here, the dynamic model is defined as a mathematical representation of the relation between bus voltage, bus frequency, and active/reactive power flowing in the bus. With this form of load representation, Long-Short Term Memory (LSTM), a variation of RNN, performed better than the concurrent algorithms like Support Vector Regression (SVR). The result of this study is a load model consisting of parameters defining the load at load bus whose predictions are compared against simulated parameters to examine their validity for use in contingency analysis. / Master of Science / Independent system Operators (ISO) and Distribution system operators (DSO) have a responsibility to provide uninterrupted power supply to consumers. That along with the longing to keep operating cost minimum, engineers and planners study the system beforehand and seek to find the optimum capacity for each of the power system elements like generators, transformers, transmission lines, etc. Then they test the overall system using power system models, which are mathematical representation of the real components, to verify the stability and strength of the system. However, the verification is only as good as the system models that are used. As most of the power systems components are controlled by the operators themselves, it is easy to develop a model from their perspective. The load is the only component controlled by consumers. Hence, the necessity of better load models. Several studies have been made on static load modeling and the performance is on par with real behavior. But dynamic loading, which is a load behavior dependent on time, is rather difficult to model. Some attempts on dynamic load modeling can be found already. Physical component-based and mathematical transfer function based dynamic models are quite widely used for the study. These load structures are largely accepted as a good representation of the systems dynamic behavior. With a load structure in hand, the next task is estimating their parameters. In this research, we tested out some new machine learning methods to accurately estimate the parameters. Thousands of simulated data are used to train machine learning models. After training, we validated the models on some other unseen data. This study finally goes on to recommend better methods to load modeling.

Yes / Cyber-aggression is an offensive behaviour attacking people based on race, ethnicity, religion, gender, sexual orientation, and other traits. It has become a major issue plaguing the online social media. In this research, we have developed a deep learning-based model to identify different levels of aggression (direct, indirect and no aggression) in a social media post in a bilingual scenario. The model is an autoencoder built using the LSTM network and trained with non-aggressive comments only. Any aggressive comment (direct or indirect) will be regarded as an anomaly to the system and will be marked as Overtly (direct) or Covertly (indirect) aggressive comment depending on the reconstruction loss by the autoencoder. The validation results on the dataset from two popular social media sites: Facebook and Twitter with bilingual (English and Hindi) data outperformed the current state-of-the-art models with improvements of more than 11% on the test sets of the English dataset and more than 6% on the test sets of the Hindi dataset.

[pt] A tarefa de Processamento de Linguagem Natural consiste em analisar linguagens naturais de forma computacional, facilitando o desenvolvimento de programas capazes de utilizar dados falados ou escritos. Uma das tarefas mais importantes deste campo é a Análise de Dependência. Tal tarefa consiste em analisar a estrutura gramatical de frases visando extrair aprender dados sobre suas relações de dependência. Em uma sentença, essas relações se apresentam em formato de árvore, onde todas as palavras são interdependentes. Devido ao seu uso em uma grande variedade de aplicações como Tradução Automática e Identificação de Papéis Semânticos, diversas pesquisas com diferentes abordagens são feitas nessa área visando melhorar a acurácia das árvores previstas. Uma das abordagens em questão consiste em encarar o problema como uma tarefa de classificação de tokens e dividi-la em três classificadores diferentes, um para cada sub-tarefa, e depois juntar seus resultados de forma incremental. As sub-tarefas consistem em classificar, para cada par de palavras que possuam relação paidependente, a classe gramatical do pai, a posição relativa entre os dois e a distância relativa entre as palavras. Porém, observando pesquisas anteriores nessa abordagem, notamos que o gargalo está na terceira sub-tarefa, a predição da distância entre os tokens. Redes Neurais Recorrentes são modelos que nos permitem trabalhar utilizando sequências de vetores, tornando viáveis problemas de classificação onde tanto a entrada quanto a saída do problema são sequenciais, fazendo delas uma escolha natural para o problema. Esse trabalho utiliza-se de Redes Neurais Recorrentes, em específico Long Short-Term Memory, para realizar a tarefa de predição da distância entre palavras que possuam relações de dependência como um problema de classificação sequence-to-sequence. Para sua avaliação empírica, este trabalho segue a linha de pesquisas anteriores e utiliza os dados do corpus em português disponibilizado pela Conference on Computational Natural Language Learning 2006 Shared Task. O modelo resultante alcança 95.27 por cento de precisão, resultado que é melhor do que o obtido por pesquisas feitas anteriormente para o modelo incremental. / [en] The Natural Language Processing task consists of analyzing the grammatical structure of a sentence written in natural language aiming to learn, identify and extract information related to its dependency structure. This data can be structured like a tree, since every word in a sentence has a head-dependent relation to another word from the same sentence. Since Dependency Parsing is used in many applications like Machine Translation, Semantic Role Labeling and Part-Of-Speech Tagging, researchers aiming to improve the accuracy on their models are approaching this task in many different ways. One of the approaches consists in looking at this task as a token classification problem, using different classifiers for each sub-task and joining them in an incremental way. These sub-tasks consist in classifying, for each head-dependent pair, the Part-Of-Speech tag of the head, the relative position between the two words and the distance between them. However, previous researches using this approach show that the bottleneck lies in the distance classifier. Recurrent Neural Networks are a kind of Neural Network that allows us to work using sequences of vectors, allowing for classification problems where both our input and output are sequences, making them a great choice for the problem at hand. This work studies the use of Recurrent Neural Networks, in specific Long Short-Term Memory networks, for the head-dependent distance classifier sub-task as a sequence-to-sequence classification problem. To evaluate its efficiency, this work follows the line of previous researches and makes use of the Portuguese corpus of the Conference on Computational Natural Language Learning 2006 Shared Task. The resulting model attains 95.27 percent precision, which is better than the previous results obtained using incremental models.

[pt] Prever emprego e desemprego é de grande importância para praticamente todos os agentes de uma economia. Emprego é uma das principais variáveis analisadas como indicador econômico, e desemprego serve para os policy makers como uma orientação às suas decisões. Neste trabalho, eu estudo quais características das duas séries podemos usar para auxiliar no tratamento dos dados e métodos empregados para auxiliar no poder preditivo das mesmas. Eu comparo modelos de machine (Random Forest e Lasso Adaptativo) e Deep (Long short Term memory) learning, procurando capturar as não linearidades e dinâmicas de ambas séries. Os resultados encontrados sugerem que o modelo AR com Random Forest aplicado nos resíduos, como uma maneira de separar parte linear e não linear, é o melhor modelo para previsão de emprego, enquanto Random Forest e AdaLasso com Random Forest aplicado nos resíduos são os melhores para o desemprego. / [en] Forecasting employment and unemployment is of great importance for virtually all agents in the economy. Employment is one of the main variables analyzed as an economic indicator, and unemployment serves to policy makers as a guide to their actions. In this essay, I study what features of both series we can use on data treatment and methods used to add to the forecasting predictive power. Using an AR model as a benchmark, I compare machine (Random Forest and Adaptive Lasso) and deep (Long Short Term Memory) learning methods, seeking to capture non-linearities of both series dynamics. The results suggests that an AR model with a Random Forest on residuals (as a way to separate linear and non-linear part) is the best model for employment forecast, while Random Forest and AdaLasso with Random Forest on residuals were the best for unemployment forecast.

Human activity data is typically represented as time series data, and RNNs, often with LSTM cells, are commonly used for recognition in this field. However, RNNs and LSTM-RNNs are often too resource-intensive for real-time applications on resource constrained devices, making them unsuitable. This thesis project is carried out at Wrlds AB, Stockholm. At Wrlds, all machine learning is run in the cloud, but they have been attempting to run their AI algorithms on their embedded devices. The main task of this project was to investigate alternative network structures to minimize the size of the networks to be used on human activity data. This thesis investigates the use of Fast GRNN, a deep learning algorithm developed by Microsoft researchers, to classify human activity on resource-constrained devices. The FastGRNN algorithm was compared to state-of-the-art RNNs, LSTM, GRU, and Simple RNN in terms of accuracy, classification time, memory usage, and energy consumption. This research is limited to implementing the FastRNN algorithm on Nordic SoCs using their SDK and TensorFlow Lite Micro. The result of this thesis shows that the proposed network has similar performance as LSTM networks in terms of accuracy while being both considerably smaller and faster, making it a promising solution for human activity recognition on embedded devices with limited computational resources and merits further investigation. / Rörelse igenkännings analys är oftast representerat av tidsseriedata där ett RNN modell meden LSTM arkitektur är oftast den självklara vägen att ta. Dock så är denna arkitektur väldigt resurskrävande för applikationer i realtid och gör att det uppstår problem med resursbegränsad hårdvara. Detta examensarbete är utfört i samarbete med Wrlds Technologies AB. På Wrlds så körs deras maskin inlärningsmodeller på molnet och lokalt på mobiltelefoner. Wrlds har nu påbörjat en resa för att kunna köra modeller direkt på små inbyggda system. Examensarbete kommer att utvärdera en FastGRNN som är en NN-arkitektur utvecklad av Microsoft i syfte att användas på resurs begränsad hårdvara. FastGRNN algoritmen jämfördes med andra högkvalitativa arkitekturer som RNNs, LSTM, GRU och en simpel RNN. Träffsäkerhet, klassifikationstid, minnesanvändning samt energikonsumtion användes för att jämföra dom olika varianterna. Detta arbete kommer bara att utvärdera en FastGRNN algoritm på en Nordic SoCs och kommer att användas deras SDK samt Tensorflow Lite Micro. Resultatet från detta examensarbete visar att det utvärderade nätverket har liknande prestanda som ett LSTM nätverk men också att nätverket är betydligt mindre i storlek och därmed snabbare. Detta betyder att ett FastGRNN visar lovande resultat för användningen av rörelseigenkänning på inbyggda system med begränsad prestanda kapacitet.

This study aims to investigate the implementation of an AI model that predicts customer purchases, in the telecom industry. The thesis also outlines how such an AI model can assist decision-making in marketing strategies. It is concluded that designing the AI model by following a Recurrent Neural Network (RNN) architecture with a Long Short-Term Memory (LSTM) layer, allow for a successful implementation with satisfactory model performances. Stepwise instructions to construct such model is presented in the methodology section of the study. The RNN-LSTM model further serves as an assisting tool for marketers to assess how a consumer’s website behavior affect their purchase behavior over time, in a quantitative way - by observing what the authors refer to as the Customer Purchase Propensity Journey (CPPJ). The firm empirical basis of CPPJ, can help organizations improve their allocation of marketing resources, as well as benefit the organization’s online presence by allowing for personalization of the customer experience. / Denna studie undersöker implementeringen av en AI-modell som förutspår kunders köp, inom telekombranschen. Studien syftar även till att påvisa hur en sådan AI-modell kan understödja beslutsfattande i marknadsföringsstrategier. Genom att designa AI-modellen med en Recurrent Neural Network (RNN) arkitektur med ett Long Short-Term Memory (LSTM) lager, drar studien slutsatsen att en sådan design möjliggör en framgångsrik implementering med tillfredsställande modellprestation. Instruktioner erhålls stegvis för att konstruera modellen i studiens metodikavsnitt. RNN-LSTM-modellen kan med fördel användas som ett hjälpande verktyg till marknadsförare för att bedöma hur en kunds beteendemönster på en hemsida påverkar deras köpbeteende över tiden, på ett kvantitativt sätt - genom att observera det ramverk som författarna kallar för Kundköpbenägenhetsresan, på engelska Customer Purchase Propensity Journey (CPPJ). Den empiriska grunden av CPPJ kan hjälpa organisationer att förbättra allokeringen av marknadsföringsresurser, samt gynna deras digitala närvaro genom att möjliggöra mer relevant personalisering i kundupplevelsen.

This thesis investigates the effectiveness of employing pretraining and a discrete "control signal" bottleneck layer in a neural network trained in aircraft navigation through deep reinforcement learning. The study defines two distinct tasks to assess the efficacy of this approach. The first task is utilized for pretraining specific parts of the network, while the second task evaluates the potential benefits of this technique. The experimental findings indicate that the network successfully learned three main macro actions during pretraining. flying straight ahead, turning left, and turning right, and achieved high rewards on the task. However, utilizing the pretrained network on the transfer task yielded poor performance, possibly due to the limited effective action space or deficiencies in the training process. The study discusses several potential solutions, such as incorporating multiple pretraining tasks and alterations of the training process as avenues for future research. Overall, this study highlights the challanges and opportunities associated with combining pretraining with a discrete bottleneck layer in the context of simulated aircraft navigation using reinforcement learning. / Denna studie undersöker effektiviteten av att använda förträning och en diskret "styrsignal" som fungerar som flaskhals i ett neuralt nätverk tränat i flygnavigering med hjälp av djup förstärkande inlärning. Studien definierar två olika uppgifter för att bedöma effektiviteten hos denna metod. Den första uppgiften används för att förträna specifika delar at nätverket, medan den andra uppgiften utvärderar de potentiella fördelarna med denna teknik. De experimentella resultaten indikerar att nätverket framgångsrikt lärde sig tre huvudsakliga makrohandlingar under förträningen: att flyga rakt fram, att svänga vänster och att svänga höger, och uppnådde höga belöningar för uppgiften. Men att använda det förtränade nätverket för den uppföljande uppgiften gav dålig prestation, möjligen på grund av det begränsade effektiva handlingsutrymmet eller begränsningar i träningsprocessen. Studien diskuterar flera potentiella lösningar, såsom att inkorporera flera förträningsuppgifter och ändringar i träningsprocessen, som möjliga framtida forskningsvägar. Sammantaget belyser denna studie de utmaningar och möjligheter som är förknippade med att kombinera förträning med ett diskret flaskhalslager inom kontexten av simulerad flygnavigering och förstärkningsinlärning.

Negation is a common property of languages, in that there are few languages, if any, that lack means to revert the truth-value of a statement. A challenge to cross-lingual studies of negation lies in the fact that languages encode and use it in different ways. Although this variation has been extensively researched in linguistics, little has been done in automated language processing. In particular, we lack computational models of processing negation that can be generalized across language. We even lack knowledge of what the development of such models would require. These models however exist and can be built by means of existing cross-lingual resources, even when annotated data for a language other than English is not available. This thesis shows this in the context of detecting string-level negation scope, i.e. the set of tokens in a sentence whose meaning is affected by a negation marker (e.g. 'not'). Our contribution has two parts. First, we investigate the scenario where annotated training data is available. We show that Bi-directional Long Short Term Memory (BiLSTM) networks are state-of-the-art models whose features can be generalized across language. We also show that these models suffer from genre effects and that for most of the corpora we have experimented with, high performance is simply an artifact of the annotation styles, where negation scope is often a span of text delimited by punctuation. Second, we investigate the scenario where annotated data is available in only one language, experimenting with model transfer. To test our approach, we first build NEGPAR, a parallel corpus annotated for negation, where pre-existing annotations on English sentences have been edited and extended to Chinese translations. We then show that transferring a model for negation scope detection across languages is possible by means of structured neural models where negation scope is detected on top of a cross-linguistically consistent representation, Universal Dependencies. On the other hand, we found cross-lingual lexical information only to help very little with performance. Finally, error analysis shows that performance is better when a negation marker is in the same dependency substructure as its scope and that some of the phenomena related to negation scope requiring lexical knowledge are still not captured correctly. In the conclusions, we tie up the contributions of this thesis and we point future work towards representing negation scope across languages at the level of logical form as well.

Sequences are ubiquitous in the domain of biology. One of the current best machine learning techniques for analysing sequences is the long short-term memory (LSTM) network. Owing to significant barriers to adoption in biology, focussed efforts are required to realize the use of LSTMs in practice. Thus, the aim of this work is to improve the state of LSTMs for biology, and we focus on biological tasks pertaining to physiological signals, peripheral neural signals, and molecules. This goal drives the three subplots in this thesis: biological applications, visualizations, and extensions. We start by demonstrating the utility of LSTMs for biological applications. On two new physiological-signal datasets, LSTMs were found to outperform hidden Markov models. LSTM-based models, implemented by other researchers, also constituted the majority of the best performing approaches on publicly available medical datasets. However, even if these models achieve the best performance on such datasets, their adoption will be limited if they fail to indicate when they are likely mistaken. Thus, we demonstrate on medical data that it is straightforward to use LSTMs in a Bayesian framework via dropout, providing model predictions with corresponding uncertainty estimates. Another dataset used to show the utility of LSTMs is a novel collection of peripheral neural signals. Manual labelling of this dataset is prohibitively expensive, and as a remedy, we propose a sequence-to-sequence model regularized by Wasserstein adversarial networks. The results indicate that the proposed model is able to infer which actions a subject performed based on its peripheral neural signals with reasonable accuracy. As these LSTMs achieve state-of-the-art performance on many biological datasets, one of the main concerns for their practical adoption is their interpretability. We explore various visualization techniques for LSTMs applied to continuous-valued medical time series and find that learning a mask to optimally delete information in the input provides useful interpretations. Furthermore, we find that the input features looked for by the LSTM align well with medical theory. For many applications, extensions of the LSTM can provide enhanced suitability. One such application is drug discovery -- another important aspect of biology. Deep learning can aid drug discovery by means of generative models, but they often produce invalid molecules due to their complex discrete structures. As a solution, we propose a version of active learning that leverages the sequential nature of the LSTM along with its Bayesian capabilities. This approach enables efficient learning of the grammar that governs the generation of discrete-valued sequences such as molecules. Efficiency is achieved by reducing the search space from one over sequences to one over the set of possible elements at each time step -- a much smaller space. Having demonstrated the suitability of LSTMs for biological applications, we seek a hardware efficient implementation. Given the success of the gated recurrent unit (GRU), which has two gates, a natural question is whether any of the LSTM gates are redundant. Research has shown that the forget gate is one of the most important gates in the LSTM. Hence, we propose a forget-gate-only version of the LSTM -- the JANET -- which outperforms both the LSTM and some of the best contemporary models on benchmark datasets, while also reducing computational cost.