Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 50 for “"Resnet-18"”.
-
Variability-aware training and self-tuning of highly quantized DNNs for analog PIM
… the gain in accuracy is up to 35.7% for ResNet-18 over the best alternative. We demonstrate that, under a realistic pattern of within- and between-chip components of variability, training alone is unable to prevent large DNN accuracy loss (of up to 54% on CIFAR- 100/ResNet-18). We …
-
Elektroencefalogramų analizė konvoliuciniais neuroniniais tinklais /
… Hat. Two CNN architectures, SqueezeNet and ResNet-18, were evaluated to determine the optimal baseline model. The combination of ResNet-18 and the complex Morlet wavelet was chosen for further improvement using the SimCLR pretraining framework and data augmentation. To enhance …
-
Inference neuronové sítě na zařízení ZYNQ
… analýza odolnosti vůči poruchám pomocí sítě ResNet-18.
-
Image-based classification of hoarding clutter using deep learning
… CIR scoring accuracy by 1% point compared to a ResNet-18-based method previously developed by Tezcan et al. Secondly, a Vision Transformer (ViT) architecture is adopted for CIR scoring, resulting in additional 5% points improvement in accuracy over ResNet-18. Thirdly, in order to further address …
-
Detekce povrchových defektů kovových dílů pomocí neuronových sítí
… AD se ukázala jako nejpřesnější architektura ResNet-18 s přesností 84,45 AUROC, pro materiály to byla architektura EfficientNet s průměrnou přesností 87,22 AUROC. Pro množiny MPDD a MPDD2 byla naměřena jako nejpřesnější architektura ResNet50 s průměrnou přesností 88,64 a 61,10 AUROC. Podle …
-
Robustness of ai-driven histopathology under real-world adversarial examples
… cohort, four representative architectures—ResNet-18, ResNet-50, HIPT-2MLP, and ViT-B/16 —are benchmarked across controlled pixel-level distortions and artifact-like transformations. Adversarial methods include iterative Fast Gradient Sign, Projected Gradient Descent, Salt-and-Pepper noise, …
-
Cut-in Detection from Videos
… of 85%. We achieve this result by employing ResNet-18 and CNN-LSTM with a tuned level of context around the target vehicle. Further we predict the estimates of interest such as start and end of cut-in intention using the same architectures and discuss the challenges.
-
Correlating microscopy and ToF-SIMS images to cellulose size using deep learning
… microscopy images were subjected to a pretrained resnet-18 image model to identify differences in chemical properties between spots. After suitable preprocessing, models could successfully be built to tell the difference between spots from early and late mass fractions. The cellulose fractions …
-
Optická kontrola waferů mikro čoček
… konvoluční neuronové sítě učenou s učitelem (ResNet-18) a algoritmus pro detekci anomálií učený bez učitele využívající vektorové vyhledávání (FAISS). V rámci řešení byl sestaven hardwarový snímací systém a vytvořena reálná průmyslová datová sada obsahující 1044 anotovaných snímků. Výsledky …
-
Self-Supervised & Explanation Based Handwriting Verification
… AND handwriting dataset. Our results show that a ResNet-based Variational Auto-Encoder (VAE) achieved 76.3% accuracy, outperforming other generative models. Additionally, a ResNet-18 model fine-tuned with Variance-Invariance-Covariance Regularization (VICReg) achieved 78% accuracy, outperforming …
-
Energy-aware DNN Quantization for Processing-In-Memory Architecture
… of EGQ on several DNN models VGG-19, ResNet-18, ResNet-50, MobileNet-V2, and SqueezeNet. Also, the area, dynamic energy, and energy efficiency in the compressed models with various memory technologies are analyzed. EGQ shows 15%-103% higher energy efficiency with 2% accuracy loss than …
-
Towards Reliable Image Classification: A Systematic Robustness Analysis of CNN and Classical Models Under Natural Corruptions
… Random Forest, MLP) to deep learning (Lenet-5, ResNet-18)—on MNIST and Fashion-MNIST. We propose a unified, multi-metric framework assessing accuracy, robustness (flip rate, label variation), uncertainty (Gini index, max probability), and efficiency (parameter count, training time) under clean, …
-
Adding noise to the momentum term in ANN to reduce overfitting
… using various architectures like MLP, CNN, and ResNet-18. The results show that adding fBm noise improves the generalization of the models compared to the original optimizers. Specifically, the experiments demonstrated that certain values of the Hurst parameter (H) in the fBm noise led to better …
-
Physicochemical Hydrodynamics and Machine Learning Characterization of Isothermal Nucleic Acid Amplification Nucleation Site Analysis
… I describe a machine learning approach using a ResNet-18 model to analyze temporal ANSA data and predict DNA concentrations. Two models classify DNA by clinical groups or log-fold changes. This work supports robust, POC-suitable HIV diagnostics and establishes a platform for broader quantitative …
-
Research on fashion classifier algorithms /
… across five architectures: LeNet-5, AlexNet, ResNet-18, EfficientNet-B0, and Vision Transformer (ViT-B/16), comparing both models trained from scratch and those utilizing transfer learning with ImageNet pre-trained weights. All models were trained on 50,000 images with similar, optimizing …
-
Fruit Detection Using CenterNet
… backbones were introduced, the backbones are ResNet-18, DLA-34, and Hourglass. A fruit dataset with four classes and 1,690 images was collected for this research work. By comparing those models with different backbone, according to the results, the deep learning-based model with DLA-34 was …
-
Breast Cancer Classification from Histopathological Images Using Transfer Learning and Deep Neural Networks
… learning from pre-trained deep neural networks ResNet18, Inception-V3Net, and ShuffleNet in terms of binary classification and multiclass classification for breast cancer from histopathological images. We use transfer learning with the fine-tuned network results in much faster and less …
-
Hardware software co-design of machine learning accelerators using univariate functions
… model, the fully parallel-pipeline HBUNN-based ResNet-18 has a reduced area cost by 29.6% compared to conventional binary architectures. Next, we introduce Personal Self-Attention (PSA), a novel method for learning non- linear univariate functions. Using PSA with linear transformations, we …
-
Autonom körning i realtid
… Efter att ha jämfört fyra olika nätverk var det ResNet-18 som var mest effektiv på minst mängd data och klarade därför att köra framåt, vänster, höger samt stanna vid hinder runt en bana. Detta arbete utfördes som ett komplementerande projekt åt företaget Knightecs tidigare examensarbete inom …
-
Rare disease classification from facial photographs using deep learning
… convolutional neural network models (AlexNet, ResNet-18, ResNet-34, ResNet-50, VGG-16, VGG-19, DenseNet121, MobileNetV2) are analyzed over two different transfer learning approaches. All the models, except VGG-16, achieved superior results when trained with fine-tuned transfer learning approach …
Page 1 of 3