mlsanityData to 7 Oct 2026

Non-Determinism in TensorFlow ResNets

arXiv 2001.11396 ↗First posted 30 Jan 2020cs.LG cs.NE stat.MLcited by 11

Not on the map. Why?

Abstract (arXiv)

We show that the stochasticity in training ResNets for image classification on GPUs in TensorFlow is dominated by the non-determinism from GPUs, rather than by the initialisation of the weights and biases of the network or by the sequence of minibatches given. The standard deviation of test set accuracy is 0.02 with fixed seeds, compared to 0.027 with different seeds---nearly 74% of the standard deviation of a ResNet model is non-deterministic. For test set loss the ratio of standard deviations is more than 80%. These results call for more robust evaluation strategies of deep learning models, as a significant amount of the variation in results across runs can arise simply from GPU randomness.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). First posted on arXiv 30 Jan 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 2; 2023: 3; 2024: 1; 2025: 0; 2026: 2.

11 papers.

  1. 2021On the Reproducibility of Neural Network Predictionscited by 28
  2. 2023Generative Models as a Complex Systems Science: How can we make sense of large language model behavior?cited by 19
  3. 2023DORIS-MAE: Scientific Document Retrieval using Multi-level Aspect-based Queriescited by 11
  4. 2023Reproducibility of Machine Learning: Terminology, Recommendations and Open Issuescited by 8
  5. 2022Engineering the Neural Automatic Passenger Countercited by 2
  6. 2024FEEL: A Framework for Evaluating Emotional Support Capability with Large Language Modelscited by 2
  7. 2021Evaluating Generic Auto-ML Tools for Computational Pathologycited by 1
  8. 2021Impact of GPU uncertainty on the training of predictive deep neural networkscited by 0
  9. 2022Incorporating Crowdsourced Annotator Distributions into Ensemble Modeling to Improve Classification Trustworthiness for Ancient Greek Papyricited by 0
  10. 2026A Model-Driven Lossless Compression Algorithm Resistant to Mismatchcited by 0
  11. 2026Omega-S: A Functional Resilience Index for LLM Fine-Tuningcited by 0