CIFAR-100

CIFAR-100 is a dataset that contains 100 classes of images, each with 600 images, used to evaluate self-supervised representation learning methods.

Papers using CIFAR-100 (18)

Expanding the Role of Diffusion Models for Robust Classifier Training2026Multivariate Variational Autoencoder2025Can Synthetic Images Conquer Forgetting? Beyond Unexplored Doubts in Few-Shot Class-Incremental Learning2025Variational Self-Supervised Learning2025Generative Adversarial Network based on Resnet for Conditional Image Restoration2017 · 10 citesInvestigating Under and Overfitting in Wasserstein Generative Adversarial Networks2019 · 8 citesData-Efficient GAN Training Beyond (Just) Augmentations: A Lottery Ticket Perspective2021 · 7 citesScore-Guided Generative Adversarial Networks2020 · 4 citesAdaptive Weighted Discriminator for Training Generative Adversarial Networks2020 · 3 citesHow good is my GAN?2018 · 2 citesPurify Unlearnable Examples via Rate-Constrained Variational Autoencoders2024 · 2 citesStructural Adversarial Objectives for Self-Supervised Representation Learning2023 · 1 citesMimicDiffusion: Purifying Adversarial Perturbation via Mimicking Clean Diffusion Model2023 · 1 citesGenerative Convolution Layer for Image Generation2021Learning Symbolic Representations Through Joint GEnerative and DIscriminative Training2023Rethinking cluster-conditioned diffusion models for label-free image synthesis2024MCGAN: Enhancing GAN Training with Regression-Based Generator Loss2024Generative Dataset Distillation Based on Diffusion Model2024