Indiana University-Purdue University Indianapolis (IUPUI) / Convolutional Neural Networks have come a long way since AlexNet. Each year the limits of the state of the art are being pushed to new levels. EfficientNet pushed the performance metrics to a new high and EfficientNetV2 even more so. Even so, architectures for mobile applications can benefit from improved accuracy and reduced model footprint. The classic Inverted Residual block has been the foundation upon which most mobile networks seek to improve. EfficientNet architecture is built using the same Inverted Residual block. In this thesis we experiment with Harmonious Bottlenecks in place of the Inverted Residuals to observe a reduction in the number of parameters and improvement in accuracy. The designed network is then deployed on the NXP i.MX 8M Mini board for Image classification. / 2023-10-11

xiii, 322 p. : col. ill. / Before Spain encountered the Americas, it first encountered the sea. This dissertation explores the roots of that encounter by examining perceptions of the sea in late medieval Castile-Leon reflected in art, literature, and law. It analyzes the changing attitudes of the Castilians towards the sea through an examination of its perceived place in their world, underscoring the complexity of Castilian attitudes toward the dangers and opportunities presented by the marine environment. Conceptual separation and union serve as the two foundational concepts employed for the analysis of evidence from each of the three genres under examination. Each genre highlights in various ways either the strong contrast drawn between land and sea or their seeming union conceptually. These complexities are manifest in a broad variety of sources, from collections of miracle tales to fifteenth century romances. Analysis of legal distinctions between land and sea reveal significant differences in perception regarding the nature of each environment and the rights and responsibilities of Castilians acting in either. Findings include that artistic sources reveal that a fearful attitude toward the sea accentuated by helplessness before its power dominated thirteenth century imagery, contrasting with the greater unity of land and sea reflected in miniatures from fifteenth century sources. A similar pattern of separation and union emerges in the literary evidence, where fear of the loss of agency when traveling at sea in early sources gives way to fifteenth century examples that praise its value. A comparison of the laws contained in the Siete Partidas with the late medieval records of the Cortes of Castile-Leon reveals that while the Castilian monarchs tended to consider the sea as firmly outside of their realm throughout the majority of the period of this study, strategic necessity led to an inexorable growth in the importance of the sea in the affairs of the kingdom generally. Together, the evidence supports the conclusion that by the mid-fourteenth century the view of the sea as other, typical of all early Castilian sources, gave way to a fifteenth century perspective that welcomed it in many respects, laying the foundation for the development of a great maritime empire. / Committee in charge: Lisa Wolverton, Chairperson; Robert Haskett, Member; David Luebke, Member; David Wacks, Outside Member

The subject of this thesis is neural network acceleration with the goal of reducing the number of floating point multiplications. The theoretical part of the thesis surveys current trends and methods used in the field of neural network acceleration. However, the focus is on the binarization techniques which allow replacing multiplications with logical operators. The theoretical base is put into practice in two ways. First is the GPU implementation of crucial binary operators in the Tensorflow framework with a performance benchmark. Second is an application of these operators in simple image classifier. Results are certainly encouraging. Implemented operators achieve speed-up by a factor of 2.5 when compared to highly optimized cuBLAS operators. The last chapter compares accuracies achieved by binarized models and their full-precision counterparts on various architectures.

Convolutional neural networks have been successfully applied to both object classification and audio classification. The aim of this thesis is to evaluate the degree of how well transfer learning of convolutional neural networks, trained in the object classification domain on large datasets (such as CIFAR-10, and ImageNet), can be applied to the audio classification domain when only a small dataset is available. In this work, four different convolutional neural networks are tested with three configurations of transfer learning against a configuration without transfer learning. This allows for testing how transfer learning and the architectural complexity of the networks affects the performance. Two of the models developed by Google (Inception-V3, Inception-ResNet-V2), are used. These models are implemented using the Keras API where they are pre-trained on the ImageNet dataset. This paper also introduces two new architectures which are developed by the authors of this thesis. These are Mini-Inception, and Mini-Inception-ResNet, and are inspired by Inception-V3 and Inception-ResNet-V2, but with a significantly lower complexity. The audio classification dataset consists of audio from RC-boats which are transformed into mel-spectrogram images. For transfer learning to be possible, Mini-Inception, and Mini-Inception-ResNet are pre-trained on the dataset CIFAR-10. The results show that transfer learning is not able to increase the performance. However, transfer learning does in some cases enable models to obtain higher performance in the earlier stages of training.

No description available.

Research work presented within this thesis propose a neoteric variant of deep convolutional neural network architecture, CondenseNeXt, designed specifically for ARM-based embedded computing platforms with constrained computational resources. CondenseNeXt is an improved version of CondenseNet, the baseline architecture whose roots can be traced back to ResNet. CondeseNeXt replaces group convolutions in CondenseNet with depthwise separable convolutions and introduces group-wise pruning, a model compression technique, to prune (remove) redundant and insignificant elements that either are irrelevant or do not affect performance of the network upon disposition. Cardinality, a new dimension to the existing spatial dimensions, and class-balanced focal loss function, a weighting factor inversely proportional to the number of samples, has been incorporated in order to relieve the harsh effects of pruning, into the design of CondenseNeXt’s algorithm. Furthermore, extensive analyses of this novel CNN architecture was performed on three benchmarking image datasets: CIFAR-10, CIFAR-100 and ImageNet by deploying the trained weight on to an ARM-based embedded computing platform: NXP BlueBox 2.0, for real-time image classification. The outputs are observed in real-time in RTMaps Remote Studio’s console to verify the correctness of classes being predicted. CondenseNeXt achieves state-of-the-art image classification performance on three benchmark datasets including CIFAR-10 (4.79% top-1 error), CIFAR-100 (21.98% top-1 error) and ImageNet (7.91% single model, single crop top-5 error), and up to 59.98% reduction in forward FLOPs compared to CondenseNet. CondenseNeXt can also achieve a final trained model size of 2.9 MB, however at the cost of 2.26% in accuracy loss. Thus, performing image classification on ARM-Based computing platforms without requiring a CUDA enabled GPU support, with outstanding efficiency.<br>

<p>Convolutional Neural Networks have come a long way since  AlexNet. Each year the limits of the state of the art are being pushed to new  levels. EfficientNet pushed the performance metrics to a new high and EfficientNetV2 even more so. Even so, architectures for mobile applications can benefit from improved accuracy and reduced model footprint. The classic Inverted Residual block has been the foundation upon which most mobile networks seek to improve. EfficientNet architecture is built using the same Inverted Residual block. In this paper we experiment with Harmonious Bottlenecks in  place of the Inverted Residuals to observe a reduction in the number of parameters and improvement in accuracy. The designed network is then deployed on the NXP i.MX 8M Mini board for Image classification.</p>