Adversarial Pseudo-replay for Exemplar-free Class-incremental Learning

• URL: http://arxiv.org/abs/2511.17973v1
• Date: Sat, 22 Nov 2025 08:20:09 GMT
• Title: Adversarial Pseudo-replay for Exemplar-free Class-incremental Learning
• Authors: Hiroto Honda,
• Abstract summary: Exemplar-free class-incremental learning (EFCIL) aims to retain old knowledge acquired in the previous task while learning new classes, without storing the previous images due to storage constraints or privacy concerns.<n>In this paper, we introduce adversarial pseudo-replay (APR), a method that perturbs the images of the new task with adversarial attack, to synthesize the pseudo-replay images online without storing any replay samples.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Exemplar-free class-incremental learning (EFCIL) aims to retain old knowledge acquired in the previous task while learning new classes, without storing the previous images due to storage constraints or privacy concerns. In EFCIL, the plasticity-stability dilemma, learning new tasks versus catastrophic forgetting, is a significant challenge, primarily due to the unavailability of images from earlier tasks. In this paper, we introduce adversarial pseudo-replay (APR), a method that perturbs the images of the new task with adversarial attack, to synthesize the pseudo-replay images online without storing any replay samples. During the new task training, the adversarial attack is conducted on the new task images with augmented old class mean prototypes as targets, and the resulting images are used for knowledge distillation to prevent semantic drift. Moreover, we calibrate the covariance matrices to compensate for the semantic drift after each task, by learning a transfer matrix on the pseudo-replay samples. Our method reconciles stability and plasticity, achieving state-of-the-art on challenging cold-start settings of the standard EFCIL benchmarks.

Related papers

• InvCoSS: Inversion-driven Continual Self-supervised Learning in Medical Multi-modal Image Pre-training [7.7475546440997265]
  Continual self-supervised learning (CSSL) in medical imaging trains a foundation model sequentially.<n>We propose InvCoSS, an inversion-driven continual self-supervised learning framework for medical multi-modal image pre-training.<n>InvCoSS inverts the pre-trained self-supervised model to generate synthetic images that approximate the original training distribution.
  arXiv  Detail & Related papers  (2025-12-22T09:53:38Z)
• CEAT: Continual Expansion and Absorption Transformer for Non-Exemplar Class-Incremental Learning [34.59310641291726]
  In real-world applications, dynamic scenarios require the models to possess the capability to learn new tasks continuously without forgetting the old knowledge. We propose a new architecture, named continual expansion and absorption transformer(CEAT) The model can learn the novel knowledge by extending the expanded-fusion layers in parallel with the frozen previous parameters. To improve the learning ability of the model, we designed a novel prototype contrastive loss to reduce the overlap between old and new classes in the feature space.
  arXiv  Detail & Related papers  (2024-03-11T12:40:12Z)
• Fine-Grained Knowledge Selection and Restoration for Non-Exemplar Class Incremental Learning [64.14254712331116]
  Non-exemplar class incremental learning aims to learn both the new and old tasks without accessing any training data from the past. We propose a novel framework of fine-grained knowledge selection and restoration.
  arXiv  Detail & Related papers  (2023-12-20T02:34:11Z)
• Continual Diffusion with STAMINA: STack-And-Mask INcremental Adapters [67.28751868277611]
  Recent work has demonstrated ability to customize text-to-image diffusion models to multiple, fine-grained concepts in a sequential manner. We show that capacity to learn new tasks reaches saturation over longer sequences. We introduce a novel method, STack-And-Mask INcremental Adapters (STAMINA), which is composed of low-ranked attention-masked adapters and customized tokens.
  arXiv  Detail & Related papers  (2023-11-30T18:04:21Z)
• 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)
• DiffusePast: Diffusion-based Generative Replay for Class Incremental Semantic Segmentation [73.54038780856554]
  Class Incremental Semantic (CISS) extends the traditional segmentation task by incrementally learning newly added classes. Previous work has introduced generative replay, which involves replaying old class samples generated from a pre-trained GAN. We propose DiffusePast, a novel framework featuring a diffusion-based generative replay module that generates semantically accurate images with more reliable masks guided by different instructions.
  arXiv  Detail & Related papers  (2023-08-02T13:13:18Z)
• Task-Adaptive Saliency Guidance for Exemplar-free Class Incremental Learning [60.501201259732625]
  We introduce task-adaptive saliency for EFCIL and propose a new framework, which we call Task-Adaptive Saliency Supervision (TASS) Our experiments demonstrate that our method can better preserve saliency maps across tasks and achieve state-of-the-art results on the CIFAR-100, Tiny-ImageNet, and ImageNet-Subset EFCIL benchmarks.
  arXiv  Detail & Related papers  (2022-12-16T02:43:52Z)
• Exemplar-free Continual Learning of Vision Transformers via Gated Class-Attention and Cascaded Feature Drift Compensation [38.40290722515599]
  The main challenge of exemplar-free continual learning is maintaining plasticity of the learner without causing catastrophic forgetting of previously learned tasks. We propose a new method of feature drift compensation that accommodates feature drift in the backbone when learning new tasks.
  arXiv  Detail & Related papers  (2022-11-22T14:13:15Z)
• Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning [73.24988226158497]
  We consider the high-impact problem of Data-Free Class-Incremental Learning (DFCIL) We propose a novel incremental distillation strategy for DFCIL, contributing a modified cross-entropy training and importance-weighted feature distillation. Our method results in up to a 25.1% increase in final task accuracy (absolute difference) compared to SOTA DFCIL methods for common class-incremental benchmarks.
  arXiv  Detail & Related papers  (2021-06-17T17:56:08Z)
• Memory-Efficient Incremental Learning Through Feature Adaptation [71.1449769528535]
  We introduce an approach for incremental learning that preserves feature descriptors of training images from previously learned classes. Keeping the much lower-dimensional feature embeddings of images reduces the memory footprint significantly. Experimental results show that our method achieves state-of-the-art classification accuracy in incremental learning benchmarks.
  arXiv  Detail & Related papers  (2020-04-01T21:16:05Z)

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.