Optimizing Reusable Knowledge for Continual Learning via Metalearning

2021-06-09 · NeurIPS 2021 12 · Julio Hurtado, Alain Raymond-Saez, Alvaro Soto

When learning tasks over time, artificial neural networks suffer from a problem known as Catastrophic Forgetting (CF). This happens when the weights of a network are overwritten during the training of a new task causing forgetting of old information. To address this issue, we propose MetA Reusable Knowledge or MARK, a new method that fosters weight reusability instead of overwriting when learning a new task. Specifically, MARK keeps a set of shared weights among tasks. We envision these shared weights as a common Knowledge Base (KB) that is not only used to learn new tasks, but also enriched with new knowledge as the model learns new tasks. Key components behind MARK are two-fold. On the one hand, a metalearning approach provides the key mechanism to incrementally enrich the KB with new knowledge and to foster weight reusability among tasks. On the other hand, a set of trainable masks provides the key mechanism to selectively choose from the KB relevant weights to solve each task. By using MARK, we achieve state of the art results in several popular benchmarks, surpassing the best performing methods in terms of average accuracy by over 10% on the 20-Split-MiniImageNet dataset, while achieving almost zero forgetfulness using 55% of the number of parameters. Furthermore, an ablation study provides evidence that, indeed, MARK is learning reusable knowledge that is selectively used by each task.

📄 PDF Abstract BibTeX arXiv:2106.05390

Code (1)

JuliousHurtado/meta-training-setup 공식 구현 pytorch

Tasks

Continual Learning

Optimizing Loss Functions Through Multivariate Taylor Polynomial Parameterization

2020-01-31 · Santiago Gonzalez, Risto Miikkulainen

Metalearning of deep neural network (DNN) architectures and hyperparameters has become an increasingly important area of research. Loss functions are a type of metaknowledge that is crucial to effective training of DNNs,…

Joint Optimization of Reasoning and Dual-Memory for Self-Learning Diagnostic Agent

2026-04-08 · Bingxuan Li, Simo Du, Yue Guo arxiv

Clinical expertise improves not only by acquiring medical knowledge, but by accumulating experience that yields reusable diagnostic patterns. Recent LLMs-based diagnostic agents have shown promising progress in clinical …

Continual Learning

Metalearning for Feature Selection

2017-03-20 · Ben Goertzel, Nil Geisweiller, Chris Poulin

A general formulation of optimization problems in which various candidate solutions may use different feature-sets is presented, encompassing supervised classification, automated program learning and other cases. A novel…

Classification feature selection General Classification text-classification+1

Multilayer Neuromodulated Architectures for Memory-Constrained Online Continual Learning

2020-06-12 · ICML Workshop LifelongML 2020 7 · Sandeep Madireddy, Angel Yanguas-Gil, Prasanna Balaprakash

We focus on the problem of how to achieve online continual learning under memory-constrained conditions where the input data may not be known \emph{a priori}. These constraints are relevant in edge computing scenarios. W…

Bayesian Optimization class-incremental learning Class Incremental Learning Continual Learning+4

Reasoning Portability: Guiding Continual Learning for MLLMs in the RLVR Era

2026-05-17 · Qiuhe Hong, Yuyang Liu, Shuo Yang, Tiantian Peng 외 arxiv

Vision-Language Models in Continual Learning (VLM-CL) aim to continuously adapt to new multimodal tasks while retaining prior knowledge. The emerging paradigm that couples Multimodal Large Language Models (MLLMs) with Re…

Reinforcement Learning Continual Learning