Shot Incremental

Few-shot incremental learning (FSIL) focuses on enabling machine learning models to continuously learn new classes from limited data without forgetting previously acquired knowledge. Current research emphasizes adapting large pre-trained models, like vision and language transformers, and employing techniques such as prompt engineering, attention mechanisms, and knowledge distillation to mitigate catastrophic forgetting and overfitting. This area is crucial for building more robust and adaptable AI systems, particularly in applications with limited data availability or evolving class distributions, such as industrial quality control and personalized recommendation systems. The development of effective FSIL methods is driving progress in continual learning and improving the efficiency of AI model training.

Papers

November 11, 2024

Inductive Graph Few-shot Class Incremental Learning
Yayong Li, Peyman Moghadam, Can Peng, Nan Ye, Piotr Koniusz
Graph Neural Network Data Augmentation Incremental Learning Shot Incremental

April 2, 2024

Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learners
Keon-Hee Park, Kyungwoo Song, Gyeong-Moon Park
Knowledge Distillation Vision Paper Class Incremental Learning Language Transformer Knowledge Distillation Loss Shot Incremental

March 14, 2024

Few-Shot Class Incremental Learning with Attention-Aware Self-Adaptive Prompt
Chenxi Liu, Zhenyi Wang, Tianyi Xiong, Ruibo Chen, Yihan Wu, Junfeng Guo, Heng Huang
Class Incremental Learning Prompt Based Method Shot Incremental

September 2, 2023

Big-model Driven Few-shot Continual Learning
Ziqi Gu, Chunyan Xu, Zihan Lu, Xin Liu, Anbo Dai, Zhen Cui
Continual Learning Large Model Shot Continual Learning Shot Incremental

July 29, 2022

Conservative Generator, Progressive Discriminator: Coordination of Adversaries in Few-shot Incremental Image Synthesis
Chaerin Kong, Nojun Kwak
Shot Learning GAN Model Incremental Learning Image Synthesis Adversary Agent Prior Coordination Progressive Alignment Shot Incremental Cramer Wold Generator

July 1, 2022

Few-shot incremental learning in the context of solar cell quality inspection
Julen Balzategui, Luka Eciolaza
Context Information Defect Detection Defect Class Shot Incremental

December 23, 2021

Graph Few-shot Class-incremental Learning
Zhen Tan, Kaize Ding, Ruocheng Guo, Huan Liu
Incremental Learning Graph Model Graph Meta Learning Shot Incremental