Computer science · Paper · June 2022
Christopher Sun, Jai Sharma, Milind Maiti
Roughly 6,800 natural disasters occur annually, and this alarming number continues to grow due to climate change. Effective methods to improve natural disaster response include change detection, map alignment, and vision-aided navigation to allow for the time-efficient delivery of life-saving aid. Current software functions optimally only on near-nadir images taken around ninety degrees above ground level. The inability to generalize to oblique images increases the need to compute an image's geocentric pose, which is its spatial orientation with respect to gravity. This Deep Learning investigation presents three convolutional models to predict geocentric pose using 5,923 nadir and off-nadir red, green, and blue (RGB) satellite images of cities worldwide. Prior to the ensemble, an autoencoder is assembled to condense the 256 x 256 x 3 images to 32 x 32 x 16 latent space representations, demonstrating the ability to learn useful features from the data. The first model in the ensemble is a U-Net Fully Convolutional Network (FCN) with skip connections used to predict each image's corresponding pixel-level above-ground elevation mask. This FCN achieves a median absolute deviation of 0.335 meters and an R2 of 0.865 on test data. Afterward, the elevation masks are concatenated with the RGB images to form four-channel inputs fed into the second model in the ensemble, which predicts each image's rotation angle and magnification scale, the components of its geocentric pose. This Deep Convolutional Model achieves an R2 of 0.943 on test data, significantly outperforming previous models designed by researchers. In addition to achieving superior model performance, outlier removal was performed through supervised interpolation, and the usefulness of data features was gauged through a sensitivity analysis of elevation masks, to target future avenues of feature engineering. The high-accuracy software built in this study contributes to mapping and navigation procedures to accelerate disaster relief and save human lives.
Biology and biomedicine · Paper · June 2022
Christopher Sun, Jai Sharma, Milind Maiti
The pervasiveness of cardiovascular disease and physician misdiagnosis creates the urgent need for artificial intelligence models to improve diagnosis accuracy. The first objective of this study was to train machine learning models on publicly available data sets containing simple medical information of patients to diagnose cardiovascular disease. The Multilayer Perceptron (MLP) assembled for this task performed optimally with an F1 score of 0.8968. This prompted the creation of an open-source, automated cardiovascular disease diagnosis tool, powered by the MLP. The second objective of this study was to employ a meta-learning methodology called Local Interpretable Model-Agnostic Explanations (LIME) to understand the impact of different features on the model's diagnosis in the form of marginal probabilities. K-Means Clustering was employed to segment the data into ten clusters, after which each data example was passed through LIME. The resulting histograms depict the complex relationship between feature, cluster, and impact on diagnosis. A series of P-values with contrasting orders of magnitude shows the nuances in the MLP's understanding of patients from different clusters. The results of meta-learning analysis reveal that the most important features for cardiovascular disease diagnosis are fasting blood sugar, type of chest pain, and slope of the ST segment on an electrocardiogram. Future experiments should replicate the novel methodology introduced in this study on data sets containing more specialized medical features in order to gain practical medical insights about different types of cardiovascular disease represented by each cluster. Finally, feature engineering pathways should be explored with consideration of these results to create versatile diagnosis models not only for cardiovascular disease, but adaptable to other diseases as well.
Computer science · Paper · August 2021
Jai Sharma, Milind Maiti, Christopher Sun
Dropout Regularization, serving to reduce variance, is nearly ubiquitous in Deep Learning models. We explore the relationship between the dropout rate and model complexity by training 2,000 neural networks configured with random combinations of the dropout rate and the number of hidden units in each dense layer, on each of the three data sets we selected. The generated figures, with binary cross entropy loss and binary accuracy on the z-axis, question the common assumption that adding depth to a dense layer while increasing the dropout rate will certainly enhance performance. We also discover a complex correlation between the two hyperparameters that we proceed to quantify by building additional machine learning and Deep Learning models which predict the optimal dropout rate given some hidden units in each dense layer. Linear regression and polynomial logistic regression require the use of arbitrary thresholds to select the cost data points included in the regression and to assign the cost data points a binary classification, respectively. These machine learning models have mediocre performance because their naive nature prevented the modeling of complex decision boundaries. Turning to Deep Learning models, we build neural networks that predict the optimal dropout rate given the number of hidden units in each dense layer, the desired cost, and the desired accuracy of the model. Though, this attempt encounters a mathematical error that can be attributed to the failure of the vertical line test. The ultimate Deep Learning model is a neural network whose decision boundary represents the 2,000 previously generated data points. This final model leads us to devise a promising method for tuning hyperparameters to minimize computational expense yet maximize performance. The strategy can be applied to any model hyperparameters, with the prospect of more efficient tuning in industrial models.