A comprehensive survey of continual learning: theory, method and application
To cope with real-world dynamics, an intelligent system needs to incrementally acquire,
update, accumulate, and exploit knowledge throughout its lifetime. This ability, known as …
update, accumulate, and exploit knowledge throughout its lifetime. This ability, known as …
Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding
D Li, T Wang, J Chen, Q Ren, K Kawaguchi… - Proceedings of the AAAI …, 2024 - ojs.aaai.org
Deep neural networks are susceptible to catastrophic forgetting when trained on sequential
tasks. Various continual learning (CL) methods often rely on exemplar buffers or/and …
tasks. Various continual learning (CL) methods often rely on exemplar buffers or/and …
Complementary learning subnetworks for parameter-efficient class-incremental learning
In the scenario of class-incremental learning (CIL), deep neural networks have to adapt their
model parameters to non-stationary data distributions, eg, the emergence of new classes …
model parameters to non-stationary data distributions, eg, the emergence of new classes …
Harnessing Neural Unit Dynamics for Effective and Scalable Class-Incremental Learning
Class-incremental learning (CIL) aims to train a model to learn new classes from non-
stationary data streams without forgetting old ones. In this paper, we propose a new kind of …
stationary data streams without forgetting old ones. In this paper, we propose a new kind of …
Towards Robust Continual Learning with Bayesian Adaptive Moment Regularization
J Foster, A Brintrup - arXiv preprint arXiv:2309.08546, 2023 - arxiv.org
The pursuit of long-term autonomy mandates that robotic agents must continuously adapt to
their changing environments and learn to solve new tasks. Continual learning seeks to …
their changing environments and learn to solve new tasks. Continual learning seeks to …
HyperInterval: Hypernetwork approach to training weight interval regions in continual learning
P Krukowski, A Bielawska, K Książek… - arXiv preprint arXiv …, 2024 - arxiv.org
Recently, a new Continual Learning (CL) paradigm was presented to control catastrophic
forgetting, called Interval Continual Learning (InterContiNet), which relies on enforcing …
forgetting, called Interval Continual Learning (InterContiNet), which relies on enforcing …
Enhancing Weather Model: A Meteorological Incremental Learning Method with Long-term and Short-term Asynchronous Updating Strategy
Improving model prediction capability based on continuously collected data is one of the
biggest challenges in a mature, intelligent weather forecasting system. To tackle this …
biggest challenges in a mature, intelligent weather forecasting system. To tackle this …