NeurIPS 2021 accepted papers on Continual learning
29 of the 2,508 papers accepted at NeurIPS 2021 (1.2%) are on Continual learning, compared with 1.1% at NeurIPS 2020. They fall into 4 topics.
See NeurIPS 2021 in the interactive map
Submitting to NeurIPS? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Online and replay-based · 21 papers
- DualNet: Continual Learning, Fast and Slow (59 citations)
- Gradient-based Editing of Memory Examples for Online Task-free Continual Learning (39 citations)
- Continual World: A Robotic Benchmark For Continual Reinforcement Learning (24 citations)
- Learning where to learn: Gradient sparsity in meta and continual learning (24 citations)
- Flattening Sharpness for Dynamic Gradient Projection Memory Benefits Continual Learning (23 citations)
- AFEC: Active Forgetting of Negative Transfer in Continual Learning (18 citations)
- BNS: Building Network Structures Dynamically for Continual Learning (17 citations)
- Mitigating Forgetting in Online Continual Learning with Neuron Calibration (16 citations)
- The CLEAR Benchmark: Continual LEArning on Real-World Imagery (15 citations)
- BooVAE: Boosting Approach for Continual Learning of VAE (14 citations)
- Generative vs. Discriminative: Rethinking The Meta-Continual Learning (11 citations)
- Continual Learning via Local Module Composition (7 citations)
and 9 more in the interactive map.
Class-incremental · 5 papers
- Class-Incremental Learning via Dual Augmentation (68 citations)
- Overcoming Catastrophic Forgetting in Incremental Few-Shot Learning by Finding Flat Minima (14 citations)
- SSUL: Semantic Segmentation with Unknown Label for Exemplar-based Class-Incremental Learning (10 citations)
- Bridging Non Co-occurrence with Unlabeled In-the-wild Data for Incremental Object Detection (5 citations)
- RMM: Reinforced Memory Management for Class-Incremental Learning
