China
Africa
Explore chapters and articles related to this topic
Deep Learning Neural Networks
Published in Mark Chang, Artificial Intelligence for Drug Development, Precision Medicine, and Healthcare, 2020
The CIFAR-10 dataset (Canadian Institute For Advanced Research) is a collection of images that are commonly used to train machine learning and computer vision algorithms. It is one of the most widely used datasets for machine learning research. The CIFAR-10 dataset contains 60,000 32 × 32 color images in 10 different classes, including airplanes, cars, birds, cats, deer, dogs, frogs, horses, ships, and trucks. For these images of size 32 × 32 × 3 (32-pixel wide, 32-pixel high, 3 color channels), a single fully-connected neuron in a first hidden layer of a regular neural network would have 32 × 32 × 3 = 3072 weights. This amount might be still manageable., However, for high resolution fullscreen images (computer screen resolution of 1024 × 768), 1024 × 768 × 3 = 2,359,296 weights need to be computed. It clear that this fully-connected structure does not scale to larger images, and the huge number of hyperparameters or weights will quickly lead to overfitting. One of the solutions to the problem is the use of local filters, which will be discussed a bit later.
Deep Neural Networks (DNNs) for Images Analysis
Published in Mehdi Ghayoumi, Deep Learning in Practice, 2021
These are some available datasets. At first, load the dataset and split them into three parts (training, validation, and testing). These are the objects in the CIFAR10: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, and truck (Figure 6.15).
Energy–entropy competition and the effectiveness of stochastic gradient descent in machine learning
Published in Molecular Physics, 2018
Yao Zhang, Andrew M. Saxe, Madhu S. Advani, Alpha A. Lee
The CIFAR-10 dataset consists of 60,000 colour images of 10 different classes of objects (e.g. airplane, automobile, bird, etc.), with 6000 images per class. The original machine learning problem is to parameterise a model that can classify an unseen image into one of those 10 classes. For numerical simplicity, we consider a stripped-down version of this problem where we only consider the binary classification problem of determining whether an image is an airplane or automobile. Moreover, we down sample the images to . Those simplifications are needed because numerically computing the Hessian for industrial scale models is numerically challenging.
New convolutional neural network models for efficient object recognition with humanoid robots
Published in Journal of Information and Telecommunication, 2022
Simge Nur Aslan, Ayşegül Uçar, Cüneyt Güzeliş
The MNIST and CIFAR-10 datasets are first used for object recognition on the proposed four CNN models. The results on the MNIST dataset show that the proposed models provide the best accuracy of 99.40% even for small iteration numbers. Moreover, the models have smaller structures over those of the literature. The results on the CIFAR-10 dataset show that the proposed models provided high accuracy about 82.5% although they are the smallest parameter numbers and trained for only 20 epochs.
The transition module: a method for preventing overfitting in convolutional neural networks
Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2019
S. Akbar, M. Peikari, S. Salama, S. Nofech-Mozes, A. L. Martel
We evaluated the performance of the proposed transition module on two independent data-sets. Some example patches from each data-set are shown in Figure 3. A description of each data-set is as follows.In-House: an in-house data-set acquired from the Department of Anatomic Pathology in Sunnybrook Health Sciences Centre. This data-set comprises of 1229 image patches extracted and labelled from breast WSIs scanned at x40 objective by a Scanscope XT (Aperio technologies, Leica Biosystems) scanner. Each RGB patch of size was hand selected from 31 WSIs, each one from a single patient, by a trained pathologist. The surgical excision specimens represent sections from residual invasive and in situ breast carcinoma after presurgical systemic therapy (also known as neo-adjuvant therapy). Each image patch was confirmed to contain either malignant or benign tissue by an expert pathologist. ‘Benign’ refers to patches which are absent of cancer cells but may contain epithelial and stromal elements with normal or spectrum of benign changes (Figure 3). 5-fold cross validation was used to evaluate the performance over 100 epochs.BreaKHis: a public data-set, BreaKHis (Spanhol et al. 2016b) which contains scanned images of benign (adenosis, fibroadenoma, phyllodes tumour, tubular adenoma) and malignant (ductal carcinoma, lobular carcinoma, mucinous carcinoma, papillary carcinoma) breast tumours at 40X magnification. Images were resampled into patches of dimensions , suitable for the CNNs we adopted, resulting in 11, 800 image patches in total. BreaKHis was validated using 2-fold cross-validation and across 30 epochs.MNIST, CIFAR-10: two publically available data-sets containing greyscale 2424 pixel images of handwritten digits from 0 to 9 (MNIST), and RGB 3232 pixel images of random everyday objects such as airplanes and cats divided into 10 classes (CIFAR-10). As images in these data-sets are considerably smaller than the two data-sets described above, we opted to train LeNet-5 (LeCun et al. 1998) CNN architectures which originally has two convolution layers. Each data-set has an independent train and test set and we trained the network for 30 epochs.