paper-with-me

홈 › Papers

GEX: A flexible method for approximating influence via Geometric Ensemble

2023-09-21 · NeurIPS 2023 11

Through a deeper understanding of predictions of neural networks, Influence Function (IF) has been applied to various tasks such as detecting and relabeling mislabeled samples, dataset pruning, and separation of data sources in practice. However, we found standard approximations of IF suffer from performance degradation due to oversimplified influence distributions caused by their bilinear approximation, suppressing the expressive power of samples with a relatively strong influence. To address this issue, we propose a new interpretation of existing IF approximations as an average relationship between two linearized losses over parameters sampled from the Laplace approximation (LA). In doing so, we highlight two significant limitations of current IF approximations: the linearity of gradients and the singularity of Hessian. Accordingly, by improving each point, we introduce a new IF approximation method with the following features: i) the removal of linearization to alleviate the bilinear constraint and ii) the utilization of Geometric Ensemble (GE) tailored for non-linear losses. Empirically, our approach outperforms existing IF approximations for downstream tasks with lighter computation, thereby providing new feasibility of low-complexity/nonlinear-based IF design.

📄 PDF Abstract BibTeX

Code (1)

sungyubkim/gex 공식 구현 jax

Similar Papers 제목 키워드 기반

RADIN: Souping on a Budget

2024-01-31 · Thibaut Menes, Olivier Risser-Maroix

Model Soups, extending Stochastic Weights Averaging (SWA), combine models fine-tuned with different hyperparameters. Yet, their adoption is hindered by computational challenges due to subset selection issues. In this pap…

Ensemble LearningModel Compression

Wasserstein variational gradient descent: From semi-discrete optimal transport to ensemble variational inference

2018-11-07 · Luca Ambrogioni, Umut Guclu, Marcel van Gerven

Particle-based variational inference offers a flexible way of approximating complex posterior distributions with a set of particles. In this paper we introduce a new particle-based variational inference method based on t…

Variational Inference

Geometric Neural Diffusion Processes

2023-07-11 · NeurIPS 2023 11 · Emile Mathieu, Vincent Dutordoir, Michael J. Hutchinson, Valentin De Bortoli 외

Denoising diffusion models have proven to be a flexible and effective paradigm for generative modelling. Their recent extension to infinite dimensional Euclidean spaces has allowed for the modelling of stochastic process…

Denoising

Learning from Peers: Collaborative Ensemble Adversarial Training

2025-08-27 · Li Dengjin, Guo Yanming, Xie Yuxiang, Li Zheng 외 arxiv

Ensemble Adversarial Training (EAT) attempts to enhance the robustness of models against adversarial attacks by leveraging multiple models. However, current EAT strategies tend to train the sub-models independently, igno…

Distilling Ensembles Improves Uncertainty Estimates

2020-11-23 · pproximateinference AABI Symposium 2021 1 · Zelda E Mariet, Rodolphe Jenatton, Florian Wenzel, Dustin Tran

We seek to bridge the performance gap between batch ensembles (ensembles of deep networks with shared parameters) and deep ensembles on tasks which require not only predictions, but also uncertainty estimates for these …