Technical note on Fisher Information for Robust Federated Cross-Validation

• URL: http://arxiv.org/abs/2510.03838v1
• Date: Sat, 04 Oct 2025 15:30:04 GMT
• Title: Technical note on Fisher Information for Robust Federated Cross-Validation
• Authors: Behraj Khan, Tahir Qasim Syed,
• Abstract summary: We propose Fisher Information for Robust fEderated validation (textbfFIRE)<n>Fire outperforms importance weighting benchmarks by $5.1%$ at maximum and federated learning benchmarks by up to $5.3%$ on shifted validation sets.
• Score: 3.5808917363708743
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: When training data are fragmented across batches or federated-learned across different geographic locations, trained models manifest performance degradation. That degradation partly owes to covariate shift induced by data having been fragmented across time and space and producing dissimilar empirical training distributions. Each fragment's distribution is slightly different to a hypothetical unfragmented training distribution of covariates, and to the single validation distribution. To address this problem, we propose Fisher Information for Robust fEderated validation (\textbf{FIRE}). This method accumulates fragmentation-induced covariate shift divergences from the global training distribution via an approximate Fisher information. That term, which we prove to be a more computationally-tractable estimate, is then used as a per-fragment loss penalty, enabling scalable distribution alignment. FIRE outperforms importance weighting benchmarks by $5.1\%$ at maximum and federated learning (FL) benchmarks by up to $5.3\%$ on shifted validation sets.

Related papers

• Distributional Training Data Attribution [20.18145179467698]
  We introduce distributional training data attribution (d-TDA) to predict how the distribution of model outputs depends upon the dataset.<n>We identify training examples that drastically change the distribution of some target measurement without necessarily changing the mean.<n>We also find that influence functions (IFs) emerge naturally from our distributional framework as the limit to unrolled differentiation.
  arXiv  Detail & Related papers  (2025-06-15T21:02:36Z)
• Mitigating covariate shift in non-colocated data with learned parameter priors [0.0]
  We present textitFragmentation-induced co-shift remediation ($FIcsR$), which minimizes an $f$-divergence between a fragment's covariate distribution and that of the standard cross-validation baseline. We run extensive classification experiments on multiple data classes, over $40$ datasets, and with data batched over multiple sequence lengths. The results are promising under all these conditions; with improved accuracy against batch and fold state-of-the-art by more than $5%$ and $10%$, respectively.
  arXiv  Detail & Related papers  (2024-11-10T15:48:29Z)
• Consistent Diffusion Models: Mitigating Sampling Drift by Learning to be Consistent [97.64313409741614]
  We propose to enforce a emphconsistency property which states that predictions of the model on its own generated data are consistent across time. We show that our novel training objective yields state-of-the-art results for conditional and unconditional generation in CIFAR-10 and baseline improvements in AFHQ and FFHQ.
  arXiv  Detail & Related papers  (2023-02-17T18:45:04Z)
• Integrating Local Real Data with Global Gradient Prototypes for Classifier Re-Balancing in Federated Long-Tailed Learning [60.41501515192088]
  Federated Learning (FL) has become a popular distributed learning paradigm that involves multiple clients training a global model collaboratively. The data samples usually follow a long-tailed distribution in the real world, and FL on the decentralized and long-tailed data yields a poorly-behaved global model. In this work, we integrate the local real data with the global gradient prototypes to form the local balanced datasets.
  arXiv  Detail & Related papers  (2023-01-25T03:18:10Z)
• Learnable Distribution Calibration for Few-Shot Class-Incremental Learning [122.2241120474278]
  Few-shot class-incremental learning (FSCIL) faces challenges of memorizing old class distributions and estimating new class distributions given few training samples. We propose a learnable distribution calibration (LDC) approach, with the aim to systematically solve these two challenges using a unified framework.
  arXiv  Detail & Related papers  (2022-10-01T09:40:26Z)
• Wakeword Detection under Distribution Shifts [4.128269694687]
  We propose a novel approach for semi-supervised learning (SSL) designed to overcome distribution shifts between training and real-world data. We develop a teacher labeling strategy based on confidences to reduce entropy on the label distribution from the teacher model.
  arXiv  Detail & Related papers  (2022-07-13T17:35:08Z)
• KL Guided Domain Adaptation [88.19298405363452]
  Domain adaptation is an important problem and often needed for real-world applications. A common approach in the domain adaptation literature is to learn a representation of the input that has the same distributions over the source and the target domain. We show that with a probabilistic representation network, the KL term can be estimated efficiently via minibatch samples.
  arXiv  Detail & Related papers  (2021-06-14T22:24:23Z)
• Examining and Combating Spurious Features under Distribution Shift [94.31956965507085]
  We define and analyze robust and spurious representations using the information-theoretic concept of minimal sufficient statistics. We prove that even when there is only bias of the input distribution, models can still pick up spurious features from their training data. Inspired by our analysis, we demonstrate that group DRO can fail when groups do not directly account for various spurious correlations.
  arXiv  Detail & Related papers  (2021-06-14T05:39:09Z)
• Posterior Re-calibration for Imbalanced Datasets [33.379680556475314]
  Neural Networks can perform poorly when the training label distribution is heavily imbalanced. We derive a post-training prior rebalancing technique that can be solved through a KL-divergence based optimization. Our results on six different datasets and five different architectures show state of art accuracy.
  arXiv  Detail & Related papers  (2020-10-22T15:57:14Z)
• Learning Calibrated Uncertainties for Domain Shift: A Distributionally Robust Learning Approach [150.8920602230832]
  We propose a framework for learning calibrated uncertainties under domain shifts. In particular, the density ratio estimation reflects the closeness of a target (test) sample to the source (training) distribution. We show that our proposed method generates calibrated uncertainties that benefit downstream tasks.
  arXiv  Detail & Related papers  (2020-10-08T02:10:54Z)
• Robust Federated Learning: The Case of Affine Distribution Shifts [41.27887358989414]
  We develop a robust federated learning algorithm that achieves satisfactory performance against distribution shifts in users' samples. We show that an affine distribution shift indeed suffices to significantly decrease the performance of the learnt classifier in a new test user.
  arXiv  Detail & Related papers  (2020-06-16T03:43:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.