Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

analytic classifier

An analytic classifier is a machine learning model whose classification parameters are calculated directly through closed-form mathematical formulations rather than through iterative, gradient-based optimization algorithms. Often implemented using linear regression formulations, regularized least squares, or matrix inversion on extracted feature representations, such a classifier obtains exact and deterministic solutions. In incremental and continuous learning scenarios, this analytical framework allows the model to recursively update its weights as new classes or samples arrive, yielding solutions mathematically equivalent to training on all historical and new data simultaneously without needing to store or revisit previous data samples.

1 item

ACIL: Analytic Class-Incremental Learning with Absolute Memorization and Privacy Protection

ACIL: Analytic Class-Incremental Learning with Absolute Memorization and Privacy Protection

Huiping Zhuang, Zhenyu Weng, Hongxin Wei, Renchunzi Xie, Kar-Ann Toh, Zhiping Lin

OrganizationsNanyang Technological UniversitySouth China University of TechnologyYonsei University

Why you should read this

Proposes an analytic class-incremental learning framework that mathematically matches the performance of joint training without storing historical exemplar data, eliminating catastrophic forgetting while protecting data privacy across multi-phase learning tasks.

Class-incremental learning (CIL) learns a classification model with training data of different classes arising progressively. Existing CIL either suffers from serious accuracy loss due to catastrophic forgetting, or invades data privacy by revisiting used exemplars. Inspired by linear learning formulations, we propose an analytic class-incremental learning (ACIL) with absolute memorization of past knowledge while avoiding breaching of data privacy (i.e., without storing historical data). The absolute memorization is demonstrated in the sense that class-incremental learning using ACIL given present data would give identical results to that from its joint-learning counterpart which consumes both present and historical samples. This equality is theoretically validated. Data privacy is ensured since no historical data are involved during the learning process. Empirical validations demonstrate ACIL's competitive accuracy performance with near-identical results for various incremental task settings (e.g., 5-50 phases). This also allows ACIL to outperform the state-of-the-art methods for large-phase scenarios (e.g., 25 and 50 phases).

Added

2026-09-26