paper-with-me

Papers

Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding

2024-01-17 · Depeng Li, Tianqi Wang, Junwei Chen, Qining Ren, Kenji Kawaguchi, Zhigang Zeng

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 network expansion for balancing model stability and plasticity, which, however, compromises their practical value due to privacy and memory concerns. Instead, this paper considers a strict yet realistic setting, where the training data from previous tasks is unavailable and the model size remains relatively constant during sequential training. To achieve such desiderata, we propose a conceptually simple yet effective method that attributes forgetting to layer-wise parameter overwriting and the resulting decision boundary distortion. This is achieved by the synergy between two key components: HSIC-Bottleneck Orthogonalization (HBO) implements non-overwritten parameter updates mediated by Hilbert-Schmidt independence criterion in an orthogonal space and EquiAngular Embedding (EAE) enhances decision boundary adaptation between old and new tasks with predefined basis vectors. Extensive experiments demonstrate that our method achieves competitive accuracy performance, even with absolute superiority of zero exemplar buffer and 1.02x the base model.

📄 PDF Abstract BibTeX arXiv:2401.09067

Code (0)

등록된 구현이 없습니다.

Tasks

Continual Learning

Methods 이 논문이 사용한 방법론

BASE 설명 없음

Similar Papers 제목 키워드 기반

DualHSIC: HSIC-Bottleneck and Alignment for Continual Learning

2023-04-30 · Zifeng Wang, Zheng Zhan, Yifan Gong, Yucai Shao 외

Rehearsal-based approaches are a mainstay of continual learning (CL). They mitigate the catastrophic forgetting problem by maintaining a small fixed-size buffer with a subset of data from past tasks. While most rehearsal…

Continual Learning

The HSIC Bottleneck: Deep Learning without Back-Propagation

2019-08-05 · Wan-Duo Kurt Ma, J. P. Lewis, W. Bastiaan Kleijn

We introduce the HSIC (Hilbert-Schmidt independence criterion) bottleneck for training deep neural networks. The HSIC bottleneck is an alternative to the conventional cross-entropy loss and backpropagation that has a num…

ClassificationDeep LearningGeneral Classification

Revisiting Hilbert-Schmidt Information Bottleneck for Adversarial Robustness

2021-06-04 · NeurIPS 2021 12 · Zifeng Wang, Tong Jian, Aria Masoomi, Stratis Ioannidis 외

We investigate the HSIC (Hilbert-Schmidt independence criterion) bottleneck as a regularizer for learning an adversarially robust deep neural network classifier. In addition to the usual cross-entropy loss, we add regula…

Adversarial Robustness

"Dependency Bottleneck" in Auto-encoding Architectures: an Empirical Study

2018-02-15 · Denny Wu, Yixiu Zhao, Yao-Hung Hubert Tsai, Makoto Yamada 외

Recent works investigated the generalization properties in deep neural networks (DNNs) by studying the Information Bottleneck in DNNs. However, the mea- surement of the mutual information (MI) is often inaccurate due to …

Density Estimation

Sustainable Artificial Intelligence through Continual Learning

2021-11-17 · Andrea Cossu, Marta Ziosi, Vincenzo Lomonaco

The increasing attention on Artificial Intelligence (AI) regulation has led to the definition of a set of ethical principles grouped into the Sustainable AI framework. In this article, we identify Continual Learning, an …

Continual Learning