-
Broken Neural Scaling Laws
A smoothly broken power law functional form that accurately models and extrapolates the scaling behaviors of deep neural networks for various architectures and tasks. -
Generated Video Dataset (GVD)
A large-scale generated video benchmark dataset for network training and evaluation, comprising synthetic videos from 11 different generator models. -
Off-Policy Deep Reinforcement Learning without Exploration
The dataset used in the paper is a batch of data collected from a fixed batch of data which has already been gathered, without offering further possibility for data collection. -
muNet: Evolving Pretrained Deep Neural Networks into Scalable Auto-tuning Mul...
The proposed method uses the layers of a pretrained deep neural network as building blocks to construct an ML system that can jointly solve an arbitrary number of tasks. -
CycleAdvGAN: integration of adversarial attack and defense
The MNIST and CIFAR10 datasets are used to evaluate the Cycle-Consistent Adversarial GAN (CycleAdvGAN) for image classification. -
NIH CXR database for lung segmentation
NIH CXR database for lung segmentation, including CXR images with severe abnormal findings. -
Montgomery database for lung segmentation
Montgomery database for lung segmentation, including CXR images with small- or medium-sized abnormal findings. -
JSRT database for lung segmentation
JSRT database for lung segmentation, including CXR images with small- or medium-sized abnormal findings. -
FPDeep: Scalable Acceleration of CNN Training on Deeply-Pipelined FPGA Clusters
The dataset used in this paper is a CNN training dataset, specifically VGG-16, VGG-19, and AlexNet. -
Learning Multiple Layers of Features from Tiny Images
The CIFAR-10 dataset consists of 60,000 training images and 10,000 test images. Each image is a 32×32 color image. -
Deep Placental Vessel Segmentation for Fetoscopic Mosaicking
A placental vessel segmentation dataset for fetoscopic mosaicking, consisting of 483 manually annotated images and 6 in vivo video clips. -
Recurrent Inference Machines as inverse problem solvers for MR relaxometry
A new framework for MR relaxometry using Recurrent Inference Machines (RIMs) to perform T1 and T2 mapping.
