T-CIL: Temperature Scaling using Adversarial Perturbation for Calibration in Class-Incremental Learning

• URL: http://arxiv.org/abs/2503.22163v1
• Date: Fri, 28 Mar 2025 06:02:34 GMT
• Title: T-CIL: Temperature Scaling using Adversarial Perturbation for Calibration in Class-Incremental Learning
• Authors: Seong-Hyeon Hwang, Minsu Kim, Steven Euijong Whang,
• Abstract summary: We study model confidence calibration in class-incremental learning, where models learn from sequential tasks with different class sets.<n>Most post-hoc calibration techniques are not designed to work with the limited memories of old-task data typical in class-incremental learning.<n>We propose T-CIL, a novel temperature scaling approach for class-incremental learning without a validation set for old tasks.
• Score: 27.247270530020664
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We study model confidence calibration in class-incremental learning, where models learn from sequential tasks with different class sets. While existing works primarily focus on accuracy, maintaining calibrated confidence has been largely overlooked. Unfortunately, most post-hoc calibration techniques are not designed to work with the limited memories of old-task data typical in class-incremental learning, as retaining a sufficient validation set would be impractical. Thus, we propose T-CIL, a novel temperature scaling approach for class-incremental learning without a validation set for old tasks, that leverages adversarially perturbed exemplars from memory. Directly using exemplars is inadequate for temperature optimization, since they are already used for training. The key idea of T-CIL is to perturb exemplars more strongly for old tasks than for the new task by adjusting the perturbation direction based on feature distance, with the single magnitude determined using the new-task validation set. This strategy makes the perturbation magnitude computed from the new task also applicable to old tasks, leveraging the tendency that the accuracy of old tasks is lower than that of the new task. We empirically show that T-CIL significantly outperforms various baselines in terms of calibration on real datasets and can be integrated with existing class-incremental learning techniques with minimal impact on accuracy.

Related papers

• DATS: Distance-Aware Temperature Scaling for Calibrated Class-Incremental Learning [13.864609787260298]
  Continual Learning (CL) is gaining increasing attention for its ability to enable a single model to learn incrementally from a sequence of new classes.<n>In safety-critical applications, predictive models should also be able to reliably communicate their uncertainty in a manner - that is, with confidence scores aligned to the true frequencies of target events.<n>We propose Distance-Aware Temperature Scaling (DATS), which combines prototype-based distance estimation with distance-aware calibration to infer task proximity and assign adaptive temperatures without prior task information.
  arXiv  Detail & Related papers  (2025-09-25T13:46:56Z)
• Orthogonal Projection Subspace to Aggregate Online Prior-knowledge for Continual Test-time Adaptation [67.80294336559574]
  Continual Test Time Adaptation (CTTA) is a task that requires a source pre-trained model to continually adapt to new scenarios.<n>We propose a novel pipeline, Orthogonal Projection Subspace to aggregate online Prior-knowledge, dubbed OoPk.
  arXiv  Detail & Related papers  (2025-06-23T18:17:39Z)
• Sculpting Subspaces: Constrained Full Fine-Tuning in LLMs for Continual Learning [19.27175827358111]
  Continual learning in large language models (LLMs) is prone to catastrophic forgetting, where adapting to new tasks significantly degrades performance on previously learned ones. We propose a novel continual full fine-tuning approach leveraging adaptive singular value decomposition (SVD) We evaluate our approach extensively on standard continual learning benchmarks using both encoder-decoder (T5-Large) and decoder-only (LLaMA-2 7B) models.
  arXiv  Detail & Related papers  (2025-04-09T17:59:42Z)
• Navigating Semantic Drift in Task-Agnostic Class-Incremental Learning [51.177789437682954]
  Class-incremental learning (CIL) seeks to enable a model to sequentially learn new classes while retaining knowledge of previously learned ones. Balancing flexibility and stability remains a significant challenge, particularly when the task ID is unknown. We propose a novel semantic drift calibration method that incorporates mean shift compensation and covariance calibration.
  arXiv  Detail & Related papers  (2025-02-11T13:57:30Z)
• Adaptive Retention & Correction: Test-Time Training for Continual Learning [114.5656325514408]
  A common problem in continual learning is the classification layer's bias towards the most recent task.<n>We name our approach Adaptive Retention & Correction (ARC)<n>ARC achieves an average performance increase of 2.7% and 2.6% on the CIFAR-100 and Imagenet-R datasets.
  arXiv  Detail & Related papers  (2024-05-23T08:43:09Z)
• 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)
• Rethinking Class-incremental Learning in the Era of Large Pre-trained Models via Test-Time Adaptation [20.62749699589017]
  Class-incremental learning (CIL) is a challenging task that involves sequentially learning to categorize classes from new tasks. We propose Test-Time Adaptation for Class-Incremental Learning (TTACIL) that first fine-tunes PTMs using Adapters on the first task. Our TTACIL does not undergo any forgetting, while benefiting each task with the rich PTM features.
  arXiv  Detail & Related papers  (2023-10-17T13:06:39Z)
• Towards Robust Continual Learning with Bayesian Adaptive Moment Regularization [51.34904967046097]
  Continual learning seeks to overcome the challenge of catastrophic forgetting, where a model forgets previously learnt information. We introduce a novel prior-based method that better constrains parameter growth, reducing catastrophic forgetting. Results show that BAdam achieves state-of-the-art performance for prior-based methods on challenging single-headed class-incremental experiments.
  arXiv  Detail & Related papers  (2023-09-15T17:10:51Z)
• Continual Learning with Pretrained Backbones by Tuning in the Input Space [44.97953547553997]
  The intrinsic difficulty in adapting deep learning models to non-stationary environments limits the applicability of neural networks to real-world tasks. We propose a novel strategy to make the fine-tuning procedure more effective, by avoiding to update the pre-trained part of the network and learning not only the usual classification head, but also a set of newly-introduced learnable parameters.
  arXiv  Detail & Related papers  (2023-06-05T15:11:59Z)
• Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification [57.36281142038042]
  We present a new adaptive block called Contextual Squeeze-and-Excitation (CaSE) that adjusts a pretrained neural network on a new task to significantly improve performance. We also present a new training protocol based on Coordinate-Descent called UpperCaSE that exploits meta-trained CaSE blocks and fine-tuning routines for efficient adaptation.
  arXiv  Detail & Related papers  (2022-06-20T15:25:08Z)
• Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning [81.3514358542452]
  Few-shot in-context learning (ICL) incurs substantial computational, memory, and storage costs because it involves processing all of the training examples every time a prediction is made. parameter-efficient fine-tuning offers an alternative paradigm where a small set of parameters are trained to enable a model to perform the new task. In this paper, we rigorously compare few-shot ICL and parameter-efficient fine-tuning and demonstrate that the latter offers better accuracy as well as dramatically lower computational costs.
  arXiv  Detail & Related papers  (2022-05-11T17:10:41Z)
• Class-Incremental Learning by Knowledge Distillation with Adaptive Feature Consolidation [39.97128550414934]
  We present a novel class incremental learning approach based on deep neural networks. It continually learns new tasks with limited memory for storing examples in the previous tasks. Our algorithm is based on knowledge distillation and provides a principled way to maintain the representations of old models.
  arXiv  Detail & Related papers  (2022-04-02T16:30:04Z)

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.