-
Multi-Label Continual Learning for Medical Imaging: A Novel Benchmark
A novel benchmark for multi-label image classification in medical imaging, combining new classes and domains into a challenging scenario. -
Split CIFAR100
A variant of CIFAR-100 dataset, where the original dataset is split into 20 disjoint tasks, each consisting of 2,500 samples from 5 classes. -
Generative Kernel Continual Learning
Generative kernel continual learning dataset -
BECoTTA: Input-dependent Online Blending of Experts for Continual Test-time A...
Continual Test Time Adaptation (CTTA) is re-required to adapt efficiently to continuous unseen domains while retaining previously learned knowledge. -
CRIL Dataset
The dataset used in the Continual Robot Imitation Learning (CRIL) paper, which consists of pseudo demonstrations of learned tasks and real demonstrations of new tasks. -
splitMNIST, splitFashionMNIST, splitCIFAR
The dataset used in the paper is a continual learning benchmark, consisting of three datasets: splitMNIST, splitFashionMNIST, and splitCIFAR. -
CIFAR-100 and ImageNet-R
The dataset used in the paper is not explicitly described, but it is mentioned that the authors used the CIFAR-100 and ImageNet-R benchmarks for class-incremental continual... -
Continual World
The Continual World benchmark consists of ten realistic robotic manipulation tasks. -
Light Federated and Continual Consensus (LFedCon2)
A federated and continual learning framework for classification tasks in a society of devices -
Class-Prototype Conditional Diffusion Model for Continual Learning with Gener...
The paper proposes a Class-Prototype conditional Diffusion Model (CPDM) for continual learning with generative replay. The model uses a diffusion-based generator and a... -
Custom CIFAR-100 Dataset
The dataset used in this paper is a custom dataset created for the CIFAR-100 class-incremental continual learning task. -
Sparse Diffusion Policy: A Sparse, Reusable, and Flexible Policy for Robot Le...
The Sparse Diffusion Policy (SDP) framework, which integrates Mixture of Experts (MoE) layers into the diffusion policy. -
CULT: Continual Unsupervised Learning with Typicality-Based Environment Detec...
FashionMNIST and MNIST datasets are used for continual unsupervised learning with variational autoencoders and generative replay. -
Split MS-COCO
The dataset used in the paper is the Split MS-COCO dataset, which is a comprehensive framework for continual image captioning. -
COB: Crude Oil Benchmark datasets
Real-world time-series benchmark datasets for crude oil prices with distribution shifts -
Split CIFAR
The dataset used in this paper for Continual Learning, Catastrophic Forgetting and PCA-OGD. -
Permuted MNIST
The Permuted MNIST dataset is a variation of MNIST where new tasks of comparable difficulty to the original MNIST classification task are created by permuting the pixels of... -
Split MNIST
The dataset used in this paper for Continual Learning, Catastrophic Forgetting and PCA-OGD. -
Rotated MNIST
The Rotated MNIST dataset is a subset of the MNIST dataset with images of handwritten digits rotated by 90 degrees. -
Multimodal Parameter-Efficient Few-Shot Class Incremental Learning
Few-Shot Class Incremental Learning (FSCIL) is a challenging continual learning task, where limited training examples are available during several learning sessions.
