mlsanityData to 7 Oct 2026

On Model Stability as a Function of Random Seed

arXiv 1909.10447 ↗First posted 23 Sep 2019cs.LG cs.CL cs.NEcited by 31

Not on the map. Why?

Abstract (arXiv)

In this paper, we focus on quantifying model stability as a function of random seed by investigating the effects of the induced randomness on model performance and the robustness of the model in general. We specifically perform a controlled study on the effect of random seeds on the behaviour of attention, gradient-based and surrogate model based (LIME) interpretations. Our analysis suggests that random seeds can adversely affect the consistency of models resulting in counterfactual interpretations. We propose a technique called Aggressive Stochastic Weight Averaging (ASWA)and an extension called Norm-filtered Aggressive Stochastic Weight Averaging (NASWA) which improves the stability of models over random seeds. With our ASWA and NASWA based optimization, we are able to improve the robustness of the original model, on average reducing the standard deviation of the model's performance by 72%.

Key facts. Cited by 31 papers in the map (data through 7 Oct 2026). First posted on arXiv 23 Sep 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 5; 2023: 8; 2024: 1; 2025: 7; 2026: 4.

31 papers.

Sort:

By area:

  1. 2024Exponential Moving Average of Weights in Deep Learning: Dynamics and Benefitscited by 65
  2. 2023Elo Uncovered: Robustness and Best Practices in Language Model Evaluationcited by 58
  3. 2021Randomness In Neural Network Training: Characterizing The Impact of Toolingcited by 57
  4. 2021Nondeterminism and Instability in Neural Network Optimizationcited by 47
  5. 2021On the Reproducibility of Neural Network Predictionscited by 28
  6. 2022A Systematic Review and Replicability Study of BERT4Rec for Sequential Recommendationcited by 22
  7. 2023Revisiting the Fragility of Influence Functionscited by 20
  8. 2021Understanding the Performance of Knowledge Graph Embeddings in Drug Discoverycited by 18
  9. 2021Neural Network Weights Do Not Converge to Stationary Points: An Invariant Measure Perspectivecited by 15
  10. 2023Latent State Models of Training Dynamicscited by 15
  11. 2023FAIR-Ensemble: When Fairness Naturally Emerges From Deep Ensemblingcited by 11
  12. 2025Assessing the Macro and Micro Effects of Random Seeds on Fine-Tuning Large Language Modelscited by 10
19 more
  1. 2022Why only Micro-F1? Class Weighting of Measures for Relation Classificationcited by 8
  2. 2023Reproducibility of Machine Learning: Terminology, Recommendations and Open Issuescited by 8
  3. 2022Quantifying Inherent Randomness in Machine Learning Algorithmscited by 6
  4. 2021How Emotionally Stable is ALBERT? Testing Robustness with Stochastic Weight Averaging on a Sentiment Analysis Taskcited by 5
  5. 2022Instability in clinical risk stratification models using deep learningcited by 3
  6. 2023ALE: A Simulation-Based Active Learning Evaluation Framework for the Parameter-Driven Comparison of Query Strategies for NLPcited by 2
  7. 2025Can Hessian-Based Insights Support Fault Diagnosis in Attention-based Models?cited by 2
  8. 2022An Evaluation Study of Generative Adversarial Networks for Collaborative Filteringcited by 1
  9. 2026Link Prediction or Perdition: the Seeds of Instability in Knowledge Graph Embeddingscited by 1
  10. 2023Pseudo-Label Training and Model Inertia in Neural Machine Translationcited by 0
  11. 2023A Framework for Provably Stable and Consistent Training of Deep Feedforward Networkscited by 0
  12. 2025Beginning with You: Perceptual-Initialization Improves Vision-Language Representation and Alignmentcited by 0
  13. 2025A solution to generalized learning from small training sets found in infants repeated visual experiences of individual objectscited by 0
  14. 2025ArchISMiner: A Framework for Automatic Mining of Architectural Issue-Solution Pairs from Online Developer Communitiescited by 0
  15. 2025Sequence of Expert: Boosting Imitation Planners for Autonomous Driving through Temporal Alternationcited by 0
  16. 2025When Does Pairing Seeds Reduce Variance? Evidence from a Multi-Agent Economic Simulationcited by 0
  17. 2026Diagnostics for Individual-Level Prediction Instability in Machine Learning for Healthcarecited by 0
  18. 2026Dynamic Scaled Gradient Descent for Stable Fine-Tuning for Classificationscited by 0
  19. 2026GRAIN: Group Aggregation via Min-Norm Objectivecited by 0