<p dir="ltr">The research focuses on improving natural language processing (NLP) models by integrating the hierarchical structure of language, which is essential for understanding and generating human language. The main contributions of the study are:</p><ol><li><b>Hierarchical RNN Model:</b> Development of a deep Recurrent Neural Network model that captures both explicit and implicit hierarchical structures in language.</li><li><b>Hierarchical Attention Mechanism:</b> Use of a multi-level attention mechanism to help the model prioritize relevant information at different levels of the hierarchy.</li><li><b>Latent Indicators and Efficient Training:</b> Integration of latent indicators using the Expectation-Maximization algorithm and reduction of computational complexity with Bootstrap sampling and layered training strategies.</li><li><b>Sequence-to-Sequence Model for Translation:</b> Extension of the model to translation tasks, including a novel pre-training technique and a hierarchical decoding strategy to stabilize latent indicators during generation.</li></ol><p dir="ltr">The study claims enhanced performance in various NLP tasks with results comparable to larger models, with the added benefit of increased interpretability.</p>

No description available.

Laminar-turbulent transition has a strong impact on aerodynamic performance in many practical applications. Hence, there is a practical need for developing reliable and efficient transition prediction models, which form a critical element of the CFD process for aerospace vehicles across multiple flow regimes. This dissertation explores machine learning approaches to develop transition models using data from computations based on linear stability theory. Such data provide strong correlation with the underlying physics governed by linearized disturbance equations. In the proposed transition model, a convolutional neural network-based model encodes information from boundary layer profiles into integral quantities. Such automated feature extraction capability enables generalization of the proposed model to multiple instability mechanisms, even for those where physically defined shape factor parameters cannot be defined/determined in a consistent manner. Furthermore, sequence-to-sequence mapping is used to predict the transition location based on the mean boundary layer profiles. Such an end-to-end transition model provides a significantly simplified workflow. Although the proposed model has been analyzed for two-dimensional boundary layer flows, the embedded feature extraction capability enables their generalization to other flows as well. Neural network-based nonlinear functional approximation has also been presented in the context of transport equation-based closure models. Such models have been examined for their computational complexity and invariance properties based on the transport equation of a general scalar quantity. The data-driven approaches explored here demonstrate the potential for improved transition prediction models. / Doctor of Philosophy / Surface skin friction and aerodynamic heating caused by the flow over a body significantly increases due to the transition from laminar to turbulent flow. Hence, efficient and reliable prediction of transition onset location is a critical component of simulating fluid flows in engineering applications. Currently available transition prediction tools do not provide a good balance between computational efficiency and accuracy. This dissertation explores machine learning approach to develop efficient and reliable models for predicting transition in a significantly simplified manner. Convolutional neural network is used to extract features from the state of boundary layer flow at each location along the body. These extracted features are then processed sequentially using recurrent neural network to predict the amplification of instabilities in the flow, which is directly correlated to the onset of transition. Such an automated nature of feature extraction enables the generalization of this model to multiple transition mechanisms associated with different flow conditions and geometries. Furthermore, an end-to-end mapping from flow data to transition prediction requires no user expertise in stability theory and provides a significantly simplified workflow as compared to traditional stability-based computations. Another category of neural network-based models (known as neural operators) is also examined which can learn functional mapping from input variable field to output quantities. Such models can learn directly from data for complex set of problems, without the knowledge of underlying governing equations. Such attribute can be leveraged to develop a transition prediction model which can be integrated seamlessly in flow solvers. While further development is needed, such data-driven models demonstrate the potential for improved transition prediction models.

The automotive industry is inevitably experiencing a paradigm shift from fossil fuels to electric powertrain with significant technological breakthroughs in vehicle electrification. Emerging hybrid electric vehicles were one of the first steps towards cleaner and greener vehicles with a higher fuel economy and lower emission levels. The energy management strategy in hybrid electric vehicles determines the power flow pattern and significantly affects vehicle performance. Therefore, in this thesis, a learning-based strategy is proposed to address the energy management problem of a hybrid electric vehicle in various driving conditions. The idea of a deep recurrent neural network-based energy management strategy is proposed, developed, and evaluated. Initially, a hybrid electric vehicle model with a rule-based supervisory controller is constructed for this case study to obtain training data for the deep recurrent neural network and to evaluate the performance of the proposed energy management strategy. Secondly, due to its capabilities to remember historical data, a long short-term memory recurrent neural network is designed and trained to estimate the powertrain control variables from vehicle parameters. Extensive simulations are conducted to improve the model accuracy and ensure its generalization capability. Also, several hyper-parameters and structures are specifically tuned and debugged for this purpose. The novel proposed energy management strategy takes sequential data as input to capture the characteristics of both driver and controller behaviors and improve the estimation/prediction accuracy. The energy management controller is defined as a time-series problem, and a network predictor module is implemented in the system-level controller of the hybrid electric vehicle model. According to the simulation results, the proposed strategy and prediction model demonstrated lower fuel consumption and higher accuracy compared to other learning-based energy management strategies. / Thesis / Master of Applied Science (MASc)

In this work, we produce several prediction models for aspects of hospital emergency departments. Firstly, we demonstrate the use of a recurrent neural network to predict the rate of patient arrival at a hospital emergency department. The prediction is made on a per hour basis using date, time, calendar, and weather information. Then, we present our comparison of two prediction systems on the task of replicating the human decisions of patient admittance in a typical American emergency department. Again, a recurrent neural network (RNN) was trained to learn the task of selecting the next patient from the waiting room/queue to be admitted for treatment. Lastly, we present our attempt to produce a regression model that can predict the likelihood that a given patient will leave after waiting a specific amount of time in the emergency department’s waiting-room/queue. Such a model could be used to optimize the patient’s waiting-room/queue of an ED to minimize the likelihood of patients leaving without receiving care.

No description available.

No description available.

No description available.

No description available.

No description available.