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

keyword

incremental classifier

An incremental classifier is a machine learning model designed to continuously learn and update its knowledge from sequential streams of new data or emerging classes without needing to be retrained from scratch on previously observed data. Unlike traditional batch classifiers that require full access to the entire dataset during training, an incremental classifier adapts its decision boundaries phase by phase or instance by instance as new information arrives. A central objective of such classifiers is to maintain high predictive accuracy across both newly introduced and historical categories while mitigating catastrophic forgetting, which is the tendency of predictive models to overwrite prior knowledge when trained on new patterns. This capability makes incremental classifiers particularly valuable for dynamic environments, streaming applications, and resource-constrained settings where storing full historical datasets or repeatedly executing complete retraining cycles is computationally impractical.

1 item

GKEAL: Gaussian Kernel Embedded Analytic Learning for Few-Shot Class Incremental Task

GKEAL: Gaussian Kernel Embedded Analytic Learning for Few-Shot Class Incremental Task

Huiping Zhuang, Zhenyu Weng, Run He, Zhiping Lin, Ziqian Zeng

OrganizationsNanyang Technological UniversitySouth China University of Technology

Why you should read this

Proposes a closed-form analytic learning framework for few-shot class-incremental learning that combines Gaussian kernel embeddings with recursive least-squares updates to guarantee weight-invariant classifier solutions and prevent catastrophic forgetting.

Few-shot class incremental learning (FSCIL) aims to address catastrophic forgetting during class incremental learning in a few-shot learning setting. In this paper, we approach the FSCIL by adopting analytic learning, a technique that converts network training into linear problems. This is inspired by the fact that the recursive implementation (batch-by-batch learning) of analytic learning gives identical weights to that produced by training on the entire dataset at once. The recursive implementation and the weight-identical property highly resemble the FSCIL setting (phase-by-phase learning) and its goal of avoiding catastrophic forgetting. By bridging the FSCIL with the analytic learning, we propose a Gaussian kernel embedded analytic learning (GKEAL) for FSCIL. The key components of GKEAL include the kernel analytic module which allows the GKEAL to conduct FSCIL in a recursive manner, and the augmented feature concatenation module that balances the preference between old and new tasks especially effectively under the few-shot setting. Our experiments show that the GKEAL gives state-of-the-art performance on several benchmark datasets.

Added

2026-09-26