IF2Net: Innately Forgetting-Free Networks for Continual Learning

• URL: http://arxiv.org/abs/2306.10480v1
• Date: Sun, 18 Jun 2023 05:26:49 GMT
• Title: IF2Net: Innately Forgetting-Free Networks for Continual Learning
• Authors: Depeng Li, Tianqi Wang, Bingrong Xu, Kenji Kawaguchi, Zhigang Zeng, and Ponnuthurai Nagaratnam Suganthan
• Abstract summary: Continual learning can incrementally absorb new concepts without interfering with previously learned knowledge. Motivated by the characteristics of neural networks, we investigated how to design an Innately Forgetting-Free Network (IF2Net) IF2Net allows a single network to inherently learn unlimited mapping rules without telling task identities at test time.
• Score: 49.57495829364827
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Continual learning can incrementally absorb new concepts without interfering with previously learned knowledge. Motivated by the characteristics of neural networks, in which information is stored in weights on connections, we investigated how to design an Innately Forgetting-Free Network (IF2Net) for continual learning context. This study proposed a straightforward yet effective learning paradigm by ingeniously keeping the weights relative to each seen task untouched before and after learning a new task. We first presented the novel representation-level learning on task sequences with random weights. This technique refers to tweaking the drifted representations caused by randomization back to their separate task-optimal working states, but the involved weights are frozen and reused (opposite to well-known layer-wise updates of weights). Then, sequential decision-making without forgetting can be achieved by projecting the output weight updates into the parsimonious orthogonal space, making the adaptations not disturb old knowledge while maintaining model plasticity. IF2Net allows a single network to inherently learn unlimited mapping rules without telling task identities at test time by integrating the respective strengths of randomization and orthogonalization. We validated the effectiveness of our approach in the extensive theoretical analysis and empirical study.

Related papers

• A Unified Framework for Neural Computation and Learning Over Time [56.44910327178975]
  Hamiltonian Learning is a novel unified framework for learning with neural networks "over time" It is based on differential equations that: (i) can be integrated without the need of external software solvers; (ii) generalize the well-established notion of gradient-based learning in feed-forward and recurrent networks; (iii) open to novel perspectives.
  arXiv  Detail & Related papers  (2024-09-18T14:57:13Z)
• Towards Scalable and Versatile Weight Space Learning [51.78426981947659]
  This paper introduces the SANE approach to weight-space learning. Our method extends the idea of hyper-representations towards sequential processing of subsets of neural network weights.
  arXiv  Detail & Related papers  (2024-06-14T13:12:07Z)
• Negotiated Representations to Prevent Forgetting in Machine Learning Applications [0.0]
  Catastrophic forgetting is a significant challenge in the field of machine learning. We propose a novel method for preventing catastrophic forgetting in machine learning applications.
  arXiv  Detail & Related papers  (2023-11-30T22:43:50Z)
• Look-Ahead Selective Plasticity for Continual Learning of Visual Tasks [9.82510084910641]
  We propose a new mechanism that takes place during task boundaries, i.e., when one task finishes and another starts. We evaluate the proposed methods on benchmark computer vision datasets including CIFAR10 and TinyImagenet.
  arXiv  Detail & Related papers  (2023-11-02T22:00:23Z)
• Complementary Learning Subnetworks for Parameter-Efficient Class-Incremental Learning [40.13416912075668]
  We propose a rehearsal-free CIL approach that learns continually via the synergy between two Complementary Learning Subnetworks. Our method achieves competitive results against state-of-the-art methods, especially in accuracy gain, memory cost, training efficiency, and task-order.
  arXiv  Detail & Related papers  (2023-06-21T01:43:25Z)
• Random Weights Networks Work as Loss Prior Constraint for Image Restoration [50.80507007507757]
  We present our belief Random Weights Networks can be Acted as Loss Prior Constraint for Image Restoration'' Our belief can be directly inserted into existing networks without any training and testing computational cost. To emphasize, our main focus is to spark the realms of loss function and save their current neglected status.
  arXiv  Detail & Related papers  (2023-03-29T03:43:51Z)
• Continual Learning with Invertible Generative Models [15.705568893476947]
  Catastrophic forgetting (CF) happens whenever a neural network overwrites past knowledge while being trained on new tasks. We propose a novel method that combines the strengths of regularization and generative-based rehearsal approaches.
  arXiv  Detail & Related papers  (2022-02-11T15:28:30Z)
• Incremental Embedding Learning via Zero-Shot Translation [65.94349068508863]
  Current state-of-the-art incremental learning methods tackle catastrophic forgetting problem in traditional classification networks. We propose a novel class-incremental method for embedding network, named as zero-shot translation class-incremental method (ZSTCI) In addition, ZSTCI can easily be combined with existing regularization-based incremental learning methods to further improve performance of embedding networks.
  arXiv  Detail & Related papers  (2020-12-31T08:21:37Z)
• Unsupervised Transfer Learning for Spatiotemporal Predictive Networks [90.67309545798224]
  We study how to transfer knowledge from a zoo of unsupervisedly learned models towards another network. Our motivation is that models are expected to understand complex dynamics from different sources. Our approach yields significant improvements on three benchmarks fortemporal prediction, and benefits the target even from less relevant ones.
  arXiv  Detail & Related papers  (2020-09-24T15:40:55Z)
• Pseudo-Rehearsal for Continual Learning with Normalizing Flows [15.705568893476947]
  Catastrophic forgetting (CF) happens whenever a neural network overwrites past knowledge while being trained on new tasks. We propose a novel method that combines the strengths of regularization and generative-based rehearsal approaches.
  arXiv  Detail & Related papers  (2020-07-05T20:43:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.