AAAI 2021 accepted papers on Continual learning
20 of the 1,654 papers accepted at AAAI 2021 (1.2%) are on Continual learning, compared with 0.6% at AAAI 2020. They fall into 3 topics.
See AAAI 2021 in the interactive map
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
Class-incremental · 9 papers
- Online Class-Incremental Continual Learning with Adversarial Shapley Value (187 citations)
- Few-Shot Class-Incremental Learning via Relation Knowledge Distillation (153 citations)
- Few-Shot Lifelong Learning (120 citations)
- Continual Learning by Using Information of Each Class Holistically (38 citations)
- Class-Incremental Instance Segmentation via Multi-Teacher Networks (25 citations)
- I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting (25 citations)
- Split-and-Bridge: Adaptable Class Incremental Learning within a Single Neural Network (24 citations)
- Incremental Embedding Learning via Zero-Shot Translation (18 citations)
- DeepCollaboration: Collaborative Generative and Discriminative Models for Class Incremental Learning (10 citations)
Online and replay-based · 8 papers
- Using Hindsight to Anchor Past Knowledge in Continual Learning (193 citations)
- Overcoming Catastrophic Forgetting in Graph Neural Networks with Experience Replay (135 citations)
- Overcoming Catastrophic Forgetting in Graph Neural Networks (119 citations)
- Curriculum-Meta Learning for Order-Robust Continual Relation Extraction (69 citations)
- Continual Learning for Named Entity Recognition (66 citations)
- Gated Linear Networks (14 citations)
- Asynchronous Teacher Guided Bit-wise Hard Mining for Online Hashing (7 citations)
- Continual General Chunking Problem and SyncMap (4 citations)
