Replay-and-forget-free graph class-incremental learning: A task profiling and prompting approach

• Main Authors: NIU, Chaoxi, PANG, Guansong, CHEN, Ling, LIU, Bing
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2024
• Subjects: Databases and Information Systems; Graphics and Human Computer Interfaces
• Online Access: https://ink.library.smu.edu.sg/sis_research/9875; https://ink.library.smu.edu.sg/context/sis_research/article/10875/viewcontent/8497_Replay_and_Forget_Free_Gr__1_.pdf
• Institution: Singapore Management University

Class-incremental learning (CIL) aims to continually learn a sequence of tasks, with each task consisting of a set of unique classes. Graph CIL (GCIL) follows the same setting but needs to deal with graph tasks (e.g., node classification in a graph). The key characteristic of CIL lies in the absence...