paper-with-me

홈 › Papers

A Challenge in Reweighting Data with Bilevel Optimization

2023-10-26 · Anastasia Ivanova, Pierre Ablin

In many scenarios, one uses a large training set to train a model with the goal of performing well on a smaller testing set with a different distribution. Learning a weight for each data point of the training set is an appealing solution, as it ideally allows one to automatically learn the importance of each training point for generalization on the testing set. This task is usually formalized as a bilevel optimization problem. Classical bilevel solvers are based on a warm-start strategy where both the parameters of the models and the data weights are learned at the same time. We show that this joint dynamic may lead to sub-optimal solutions, for which the final data weights are very sparse. This finding illustrates the difficulty of data reweighting and offers a clue as to why this method is rarely used in practice.

📄 PDF Abstract BibTeX arXiv:2310.17386

Code (0)

등록된 구현이 없습니다.

Tasks

Bilevel Optimization

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

ScaleBiO: Scalable Bilevel Optimization for LLM Data Reweighting

2024-06-28 · Rui Pan, Dylan Zhang, Hanning Zhang, Xingyuan Pan 외

Bilevel optimization has shown its utility across various machine learning settings, yet most algorithms in practice require second-order information, making it challenging to scale them up. Only recently, a paradigm of …

Bilevel OptimizationInstruction FollowingMath

Unlocking Global Optimality in Bilevel Optimization: A Pilot Study

2024-08-28 · Quan Xiao, Tianyi Chen

Bilevel optimization has witnessed a resurgence of interest, driven by its critical role in trustworthy and efficient AI applications. While many recent works have established convergence to stationary points or local mi…

Bilevel OptimizationRepresentation Learning

BROS: Bias-Corrected Randomized Subspaces for Memory-Efficient Single-Loop Bilevel Optimization

2026-05-11 · Hengrui Zhang, Boao Kong, Engao Zhang, Kun Yuan arxiv

Stochastic bilevel optimization (SBO) has become a standard framework for hyperparameter learning, data reweighting, representation learning, and data-mixture optimization in deep learning. Existing exact single-loop SBO…

Representation LearningBilevel Optimization

Stability and Generalization of Bilevel Programming in Hyperparameter Optimization

2021-06-08 · NeurIPS 2021 12 · Fan Bao, Guoqiang Wu, Chongxuan Li, Jun Zhu 외

The (gradient-based) bilevel programming framework is widely used in hyperparameter optimization and has achieved excellent performance empirically. Previous theoretical work mainly focuses on its optimization properties…

Hyperparameter Optimization

Communication-Efficient Robust Federated Learning with Noisy Labels

2022-06-11 · Junyi Li, Jian Pei, Heng Huang

Federated learning (FL) is a promising privacy-preserving machine learning paradigm over distributed located data. In FL, the data is kept locally by each user. This protects the user privacy, but also makes the server d…

Bilevel OptimizationFederated LearningLearning with noisy labelsPrivacy Preserving