paper-with-me

Papers

Continual learning with hypernetworks

2019-06-03 · ICLR 2020 1 · Johannes von Oswald, Christian Henning, Benjamin F. Grewe, João Sacramento

Artificial neural networks suffer from catastrophic forgetting when they are sequentially trained on multiple tasks. To overcome this problem, we present a novel approach based on task-conditioned hypernetworks, i.e., networks that generate the weights of a target model based on task identity. Continual learning (CL) is less difficult for this class of models thanks to a simple key feature: instead of recalling the input-output relations of all previously seen data, task-conditioned hypernetworks only require rehearsing task-specific weight realizations, which can be maintained in memory using a simple regularizer. Besides achieving state-of-the-art performance on standard CL benchmarks, additional experiments on long task sequences reveal that task-conditioned hypernetworks display a very large capacity to retain previous memories. Notably, such long memory lifetimes are achieved in a compressive regime, when the number of trainable hypernetwork weights is comparable or smaller than target network size. We provide insight into the structure of low-dimensional task embedding spaces (the input space of the hypernetwork) and show that task-conditioned hypernetworks demonstrate transfer learning. Finally, forward information transfer is further supported by empirical results on a challenging CL benchmark based on the CIFAR-10/100 image datasets.

📄 PDF Abstract BibTeX arXiv:1906.00695

Code (9)

chrhenning/hypercl 공식 구현 pytorch
chrhenning/hypnettorch pytorch
gahaalt/continual-learning-overview tf
gahaalt/continual-learning-with-hypernets tf
geox-lab/cmn pytorch
gmum/hint pytorch
gmum/hyperinterval pytorch
pennfranc/hypnettorch pytorch
rvl-lab-utoronto/HyperCRL pytorch

Tasks

Continual LearningTransfer Learning

Methods 이 논문이 사용한 방법론

HyperNetwork A HyperNetwork is a network that generates weights for a main network. The behavior of the main network is the same with any usual neural network: it learns to map some raw…

Similar Papers 제목 키워드 기반

Effect of Optimizer, Initializer, and Architecture of Hypernetworks on Continual Learning from Demonstration

2023-12-31 · Sayantan Auddy, Sebastian Bergner, Justus Piater

In continual learning from demonstration (CLfD), a robot learns a sequence of real-world motion skills continually from human demonstrations. Recently, hypernetworks have been successful in solving this problem. In this …

Continual Learning

Partial Hypernetworks for Continual Learning

2023-06-19 · Hamed Hemati, Vincenzo Lomonaco, Davide Bacciu, Damian Borth

Hypernetworks mitigate forgetting in continual learning (CL) by generating task-dependent weights and penalizing weight changes at a meta-model level. Unfortunately, generating all weights is not only computationally exp…

Continual Learning

Continual Learning with Dependency Preserving Hypernetworks

2022-09-16 · Dupati Srikar Chandra, Sakshi Varshney, P. K. Srijith, Sunil Gupta

Humans learn continually throughout their lifespan by accumulating diverse knowledge and fine-tuning it for future tasks. When presented with a similar goal, neural networks suffer from catastrophic forgetting if data di…

Continual Learningimage-classificationImage Classification

Continual Model-Based Reinforcement Learning with Hypernetworks

2020-09-25 · Yizhou Huang, Kevin Xie, Homanga Bharadhwaj, Florian Shkurti

Effective planning in model-based reinforcement learning (MBRL) and model-predictive control (MPC) relies on the accuracy of the learned dynamics model. In many instances of MBRL and MPC, this model is assumed to be stat…

Continual LearningmodelModel-based Reinforcement LearningModel Predictive Control+3

Prototype Augmented Hypernetworks for Continual Learning

2025-05-12 · Neil De La Fuente, Maria Pilligua, Daniel Vidal, Albin Soutiff 외

Continual learning (CL) aims to learn a sequence of tasks without forgetting prior knowledge, but gradient updates for a new task often overwrite the weights learned earlier, causing catastrophic forgetting (CF). We prop…

Continual Learning