Reducing Per-Sample Harm in Stochastic Optimization

• URL: http://arxiv.org/abs/2607.16261v1
• Date: Sun, 28 Jun 2026 18:01:27 GMT
• Title: Reducing Per-Sample Harm in Stochastic Optimization
• Abstract summary: Moderns combine gradients from the current mini-batch with historical optimization state, such as momentum or adaptive moments.<n>We term this effect harm and formalize the parameter update as an optimization problem.<n>Because the exact formulation is intractable, we introduce a highly efficient proxy.
• Score: 1.5229257192293202
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Modern optimizers combine gradients from the current mini-batch with historical optimization state, such as momentum or adaptive moments. While highly effective, aggregating across the batch and incorporating this history can produce parameter updates that increase the loss of individual samples. We term this effect harm and formalize the parameter update as an optimization problem that explicitly minimizes the conflicting impact of both batch averaging and past optimization state on current data. Because the exact formulation is intractable, we introduce a highly efficient proxy. We first reduce the problem's dimensionality to the batch size, and then drastically cut memory and speed bottlenecks by successfully restricting the optimization to the last linear layer. This hinges on the unexpected finding that this layer alone reliably captures the second-order statistics of the per-sample gradients. The resulting surrogate problem integrates readily into standard optimizers like SGD and AdamW, and can be solved using a small number of GPU-friendly iterations. Crucially, the method exhibits favorable scaling properties, as the relative computational overhead shrinks as the model size or input grows. Experiments on image classification benchmarks confirm reduced per-sample interference and improved generalization.

Related papers

• Turning Stale Gradients into Stable Gradients: Coherent Coordinate Descent with Implicit Landscape Smoothing for Lightweight Zeroth-Order Optimization [11.723535704837266]
  Zeroth-Order (ZO) optimization is pivotal for scenarios where backpropagation is unavailable.<n>Existing methods face a stark trade-off: they are sample-inefficient (e.g., standard finite differences) or suffer from high variance due to randomized estimation.<n>In this work, we propose Coherent Coordinate Descent (CoCD), a deterministic, sample-efficient, and budget-aware ZO.
  arXiv  Detail & Related papers  (2026-05-14T04:52:24Z)
• Diff3R: Feed-forward 3D Gaussian Splatting with Uncertainty-aware Differentiable Optimization [76.38917994186733]
  Diff3R is a novel framework that bridges feed-forward prediction and test-time optimization.<n>We show that it can be seamlessly integrated into existing feed-forward 3DGS architectures for both pose-given and pose-free methods.
  arXiv  Detail & Related papers  (2026-04-01T15:40:20Z)
• Low-rank Momentum Factorization for Memory Efficient Training [13.464518325870444]
  Momentum Factorized (MoFaSGD) maintains a dynamically updated low-rank SVD representation of the first-order momentum.<n>We demonstrate MoFaSGD's effectiveness on large language model benchmarks, achieving a competitive trade-off between memory reduction (e.g. LoRA) and performance.
  arXiv  Detail & Related papers  (2025-07-10T18:04:52Z)
• AdaLomo: Low-memory Optimization with Adaptive Learning Rate [59.64965955386855]
  We introduce low-memory optimization with adaptive learning rate (AdaLomo) for large language models. AdaLomo results on par with AdamW, while significantly reducing memory requirements, thereby lowering the hardware barrier to training large language models.
  arXiv  Detail & Related papers  (2023-10-16T09:04:28Z)
• Winner-Take-All Column Row Sampling for Memory Efficient Adaptation of Language Model [89.8764435351222]
  We propose a new family of unbiased estimators called WTA-CRS, for matrix production with reduced variance. Our work provides both theoretical and experimental evidence that, in the context of tuning transformers, our proposed estimators exhibit lower variance compared to existing ones.
  arXiv  Detail & Related papers  (2023-05-24T15:52:08Z)
• Self-Tuning Stochastic Optimization with Curvature-Aware Gradient Filtering [53.523517926927894]
  We explore the use of exact per-sample Hessian-vector products and gradients to construct self-tuning quadratics. We prove that our model-based procedure converges in noisy gradient setting. This is an interesting step for constructing self-tuning quadratics.
  arXiv  Detail & Related papers  (2020-11-09T22:07:30Z)
• Stochastic Optimization with Laggard Data Pipelines [65.20044914532221]
  We show that "dataechoed" extensions of common optimization methods exhibit provable improvements over their synchronous counterparts. Specifically, we show that in convex optimization with minibatches, data echoing affords speedups on the curvature-dominated part of the convergence rate, while maintaining the optimal statistical rate.
  arXiv  Detail & Related papers  (2020-10-26T14:55:31Z)
• Balancing Rates and Variance via Adaptive Batch-Size for Stochastic Optimization Problems [120.21685755278509]
  In this work, we seek to balance the fact that attenuating step-size is required for exact convergence with the fact that constant step-size learns faster in time up to an error. Rather than fixing the minibatch the step-size at the outset, we propose to allow parameters to evolve adaptively.
  arXiv  Detail & Related papers  (2020-07-02T16:02:02Z)
• Extrapolation for Large-batch Training in Deep Learning [72.61259487233214]
  We show that a host of variations can be covered in a unified framework that we propose. We prove the convergence of this novel scheme and rigorously evaluate its empirical performance on ResNet, LSTM, and Transformer.
  arXiv  Detail & Related papers  (2020-06-10T08:22:41Z)

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.