The Rich and the Simple: On the Implicit Bias of Adam and SGD

• URL: http://arxiv.org/abs/2505.24022v2
• Date: Fri, 24 Oct 2025 07:21:56 GMT
• Title: The Rich and the Simple: On the Implicit Bias of Adam and SGD
• Authors: Bhavya Vasudeva, Jung Whan Lee, Vatsal Sharan, Mahdi Soltanolkotabi,
• Abstract summary: Adam is the de facto optimization algorithm for several deep learning applications.<n>In practice, neural networks (NNs) trained with (stochastic) gradient descent (GD) are known to exhibit simplicity bias.<n>We show that Adam is more resistant to such simplicity bias.
• Score: 26.722625797251553
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Adam is the de facto optimization algorithm for several deep learning applications, but an understanding of its implicit bias and how it differs from other algorithms, particularly standard first-order methods such as (stochastic) gradient descent (GD), remains limited. In practice, neural networks (NNs) trained with SGD are known to exhibit simplicity bias -- a tendency to find simple solutions. In contrast, we show that Adam is more resistant to such simplicity bias. First, we investigate the differences in the implicit biases of Adam and GD when training two-layer ReLU NNs on a binary classification task with Gaussian data. We find that GD exhibits a simplicity bias, resulting in a linear decision boundary with a suboptimal margin, whereas Adam leads to much richer and more diverse features, producing a nonlinear boundary that is closer to the Bayes' optimal predictor. This richer decision boundary also allows Adam to achieve higher test accuracy both in-distribution and under certain distribution shifts. We theoretically prove these results by analyzing the population gradients. Next, to corroborate our theoretical findings, we present extensive empirical results showing that this property of Adam leads to superior generalization across various datasets with spurious correlations where NNs trained with SGD are known to show simplicity bias and do not generalize well under certain distributional shifts.

Related papers

• Adam Reduces a Unique Form of Sharpness: Theoretical Insights Near the Minimizer Manifold [14.185079197889806]
  We show that Adam implicitly reduces a unique form of sharpness measure shaped by its adaptive updates, leading to qualitatively different solutions from Gradient Descent.<n>More specifically, when the loss is small, Adam wanders around the manifold of minimizers and takes semi-gradients to minimize this sharpness measure in an adaptive manner.
  arXiv  Detail & Related papers  (2025-11-04T17:58:57Z)
• Understanding the Generalization of Stochastic Gradient Adam in Learning Neural Networks [38.11287525994738]
  We present the first theoretical characterization of how affects Adam's generalization.<n>Our results reveal that while both Adam and AdamW with proper weight decay converge to poor test error solutions, their mini-batch variants can achieve near-zero test error.
  arXiv  Detail & Related papers  (2025-10-13T12:48:22Z)
• The Implicit Bias of Adam on Separable Data [27.451499849532176]
  We show that when training data are linearly separable, Adam converges towards a linear gradient that achieves diminishing learning rates. Our result shed light on the difference between Adam and (stochastic) descent from a theoretical perspective.
  arXiv  Detail & Related papers  (2024-06-15T14:39:37Z)
• Graph Out-of-Distribution Generalization via Causal Intervention [69.70137479660113]
  We introduce a conceptually simple yet principled approach for training robust graph neural networks (GNNs) under node-level distribution shifts. Our method resorts to a new learning objective derived from causal inference that coordinates an environment estimator and a mixture-of-expert GNN predictor. Our model can effectively enhance generalization with various types of distribution shifts and yield up to 27.4% accuracy improvement over state-of-the-arts on graph OOD generalization benchmarks.
  arXiv  Detail & Related papers  (2024-02-18T07:49:22Z)
• Alleviating Structural Distribution Shift in Graph Anomaly Detection [70.1022676681496]
  Graph anomaly detection (GAD) is a challenging binary classification problem. Gallon neural networks (GNNs) benefit the classification of normals from aggregating homophilous neighbors. We propose a framework to mitigate the effect of heterophilous neighbors and make them invariant.
  arXiv  Detail & Related papers  (2024-01-25T13:07:34Z)
• AdamL: A fast adaptive gradient method incorporating loss function [1.6025685183216696]
  We propose AdamL, a novel variant of the Adam, that takes into account the loss function information to attain better results. We show that AdamL achieves either the fastest convergence or the lowest objective function values when compared to Adam, EAdam, and AdaBelief. In the case of vanilla convolutional neural networks, AdamL stands out from the other Adam's variants and does not require the manual adjustment of the learning rate during the later stage of the training.
  arXiv  Detail & Related papers  (2023-12-23T16:32:29Z)
• Causality and Independence Enhancement for Biased Node Classification [56.38828085943763]
  We propose a novel Causality and Independence Enhancement (CIE) framework, applicable to various graph neural networks (GNNs) Our approach estimates causal and spurious features at the node representation level and mitigates the influence of spurious correlations. Our approach CIE not only significantly enhances the performance of GNNs but outperforms state-of-the-art debiased node classification methods.
  arXiv  Detail & Related papers  (2023-10-14T13:56:24Z)
• When Neural Networks Fail to Generalize? A Model Sensitivity Perspective [82.36758565781153]
  Domain generalization (DG) aims to train a model to perform well in unseen domains under different distributions. This paper considers a more realistic yet more challenging scenario, namely Single Domain Generalization (Single-DG) We empirically ascertain a property of a model that correlates strongly with its generalization that we coin as "model sensitivity" We propose a novel strategy of Spectral Adversarial Data Augmentation (SADA) to generate augmented images targeted at the highly sensitive frequencies.
  arXiv  Detail & Related papers  (2022-12-01T20:15:15Z)
• Discovering Invariant Rationales for Graph Neural Networks [104.61908788639052]
  Intrinsic interpretability of graph neural networks (GNNs) is to find a small subset of the input graph's features. We propose a new strategy of discovering invariant rationale (DIR) to construct intrinsically interpretable GNNs.
  arXiv  Detail & Related papers  (2022-01-30T16:43:40Z)
• Understanding the Generalization of Adam in Learning Neural Networks with Proper Regularization [118.50301177912381]
  We show that Adam can converge to different solutions of the objective with provably different errors, even with weight decay globalization. We show that if convex, and the weight decay regularization is employed, any optimization algorithms including Adam will converge to the same solution.
  arXiv  Detail & Related papers  (2021-08-25T17:58:21Z)
• Optimization Variance: Exploring Generalization Properties of DNNs [83.78477167211315]
  The test error of a deep neural network (DNN) often demonstrates double descent. We propose a novel metric, optimization variance (OV), to measure the diversity of model updates.
  arXiv  Detail & Related papers  (2021-06-03T09:34:17Z)
• Towards Practical Adam: Non-Convexity, Convergence Theory, and Mini-Batch Acceleration [12.744658958445024]
  Adam is one of the most influential adaptive algorithms for training deep neural networks. Existing approaches, such as decreasing an adaptive learning rate, adopting a big batch size, have tried to promote Adam-type algorithms to converge. We introduce an alternative easy-to-check sufficient condition, which merely depends on the parameters of historical base learning rate.
  arXiv  Detail & Related papers  (2021-01-14T06:42:29Z)
• Direction Matters: On the Implicit Bias of Stochastic Gradient Descent with Moderate Learning Rate [105.62979485062756]
  This paper attempts to characterize the particular regularization effect of SGD in the moderate learning rate regime. We show that SGD converges along the large eigenvalue directions of the data matrix, while GD goes after the small eigenvalue directions.
  arXiv  Detail & Related papers  (2020-11-04T21:07:52Z)
• Fast Learning of Graph Neural Networks with Guaranteed Generalizability: One-hidden-layer Case [93.37576644429578]
  Graph neural networks (GNNs) have made great progress recently on learning from graph-structured data in practice. We provide a theoretically-grounded generalizability analysis of GNNs with one hidden layer for both regression and binary classification problems.
  arXiv  Detail & Related papers  (2020-06-25T00:45:52Z)

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.