Research map / Continual learning / Online and replay-based
Kernel Continual Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use task-specific VRF kernel ridge regression classifiers with episodic coresets | continual learning · Permuted MNIST | Accuracy: 80.1 -> 85.5 | Helped |
| use task-specific VRF kernel ridge regression classifiers with episodic coresets | continual learning · Rotated MNIST | Accuracy: 70.8 -> 81.8 | Helped |
| use task-specific VRF kernel ridge regression classifiers with episodic coresets | continual learning · Split CIFAR100 | Accuracy: 59.9 -> 62.7 | Helped |
| use task-specific VRF kernel ridge regression classifiers with episodic coresets | continual learning · Split miniImageNet | Accuracy: 49.8 -> 53.3 | Helped |
| replace the data-driven conditional prior with an uninformative Gaussian prior (ablation) | VRF · Split CIFAR100 | Accuracy: 62.70 -> 62.46 | Hurt |
| replace VRF with an RBF kernel (ablation) | kernel continual learning · Split CIFAR100 | Accuracy: 56.86 vs 62.70 | Hurt |
| replace VRF with a linear kernel (ablation) | kernel continual learning · Split CIFAR100 | Accuracy: 60.88 vs 62.70 | Hurt |
| replace VRF with a polynomial kernel (ablation) | kernel continual learning · Split CIFAR100 | Accuracy: 60.96 vs 62.70 | Hurt |
| use varied rather than fixed numbers of classes per task (ablation) | VRF · Split CIFAR100 | Accuracy: 64.02 -> 61.00 | Hurt |
| use varied rather than fixed numbers of classes per task (ablation) | VRF · Split miniImageNet | Accuracy: 51.89 -> 53.90 | Helped |
| increase coreset size (ablation) | VRF · Rotated MNIST, Permuted MNIST, Split CIFAR100, and Split miniImageNet | Accuracy | Helped |
| increase the number of random Fourier bases (ablation) | VRF · Rotated MNIST, Permuted MNIST, Split CIFAR100, and Split miniImageNet | Accuracy | Helped |
Closest papers by meaning
- Continual Learning via Sequential Function-Space Variational Inference (ICML 2022 · similarity 0.90)
- Understanding the Dynamics of Forgetting and Generalization in Continual Learning via the Neural Tangent Kernel (ICLR 2026 · similarity 0.90)
- Continual Learning Using a Kernel-Based Method Over Foundation Models (AAAI 2025 · similarity 0.89)
- Beyond Buffer Limits: Energy-Based Data Reassembly for Continual Learning (ICML 2026 · similarity 0.89)
- STAR: Stability-Inducing Weight Perturbation for Continual Learning (ICLR 2025 · similarity 0.88)
- BiLoRA: Almost-Orthogonal Parameter Spaces for Continual Learning (CVPR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.