Harnessing Multi-Semantic Hypergraph for Few-Shot Learning

Chen, Hao; Li, Linyan; Xia, Zhenping; Lyu, Fan; Zhao, Liang; Huang, Kaizhu; Feng, Wei; Hu, Fuyuan

doi:10.1007/978-3-031-18907-4_18

Cited by 2 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The target of MOO is to find a set of optimal solutions that can simultaneously optimize multiple objectives. MOO can be applied to fields that require simultaneously optimizing multiple targets such as multi-task learning (Sener and Koltun 2018;Chen et al 2023) and recommendation systems (Geng et al 2015). In this paper, we use MOO to balance the learning of head classes and tail classes.…”

Section: Related Workmentioning

confidence: 99%

Long-Tailed Learning as Multi-Objective Optimization

Li,

Lyu,

Shang

et al. 2024

AAAI

View full text Add to dashboard Cite

Real-world data is extremely imbalanced and presents a long-tailed distribution, resulting in models biased towards classes with sufficient samples and performing poorly on rare classes. Recent methods propose to rebalance classes but they undertake the seesaw dilemma (what is increasing performance on tail classes may decrease that of head classes, and vice versa). In this paper, we argue that the seesaw dilemma is derived from the gradient imbalance of different classes, in which gradients of inappropriate classes are set to important for updating, thus prone to overcompensation or undercompensation on tail classes. To achieve ideal compensation, we formulate long-tailed recognition as a multi-objective optimization problem, which fairly respects the contributions of head and tail classes simultaneously. For efficiency, we propose a Gradient-Balancing Grouping (GBG) strategy to gather the classes with similar gradient directions, thus approximately making every update under a Pareto descent direction. Our GBG method drives classes with similar gradient directions to form a more representative gradient and provides ideal compensation to the tail classes. Moreover, we conduct extensive experiments on commonly used benchmarks in long-tailed learning and demonstrate the superiority of our method over existing SOTA methods. Our code is released at https://github.com/WickyLee1998/GBG_v1.

show abstract

Section: Related Workmentioning

confidence: 99%