mlsanityData to 14 Oct 2026

Surprisingly Strong Performance Prediction with Neural Graph Features

arXiv 2404.16551 ↗First posted 25 Apr 2024cs.LGcited by 11

Not on the map. Why?

Abstract (arXiv)

Performance prediction has been a key part of the neural architecture search (NAS) process, allowing to speed up NAS algorithms by avoiding resource-consuming network training. Although many performance predictors correlate well with ground truth performance, they require training data in the form of trained networks. Recently, zero-cost proxies have been proposed as an efficient method to estimate network performance without any training. However, they are still poorly understood, exhibit biases with network properties, and their performance is limited. Inspired by the drawbacks of zero-cost proxies, we propose neural graph features (GRAF), simple to compute properties of architectural graphs. GRAF offers fast and interpretable performance prediction while outperforming zero-cost proxies and other common encodings. In combination with other zero-cost proxies, GRAF outperforms most existing performance predictors at a fraction of the cost.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 39 of its 58 references that are in the map. First posted on arXiv 25 Apr 2024.

Builds on

39 of 58 references.

References · 39

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  3. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  4. 2018How Powerful are Graph Neural Networks?cited by 4,393
  5. 2016Adversarial Machine Learning at Scalecited by 936
  6. 2018Efficient Neural Architecture Search via Parameter Sharingcited by 837
  7. 2018SNIP: Single-shot Network Pruning based on Connection Sensitivitycited by 812
  8. 2020Pruning neural networks without any data by iteratively conserving synaptic flowcited by 479
  9. 2020Picking Winning Tickets Before Training by Preserving Gradient Flowcited by 471
  10. 2020NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Searchcited by 416
  11. 2020AutoGluon-Tabular: Robust and Accurate AutoML for Structured Datacited by 383
  12. 2019NAS-Bench-101: Towards Reproducible Neural Architecture Searchcited by 344
27 more
  1. 2015Scalable Bayesian Optimization Using Deep Neural Networkscited by 284
  2. 2020Neural Architecture Search without Trainingcited by 244
  3. 2021Zero-Cost Proxies for Lightweight NAScited by 199
  4. 2018Neural Architecture Search with Bayesian Optimisation and Optimal Transportcited by 195
  5. 2018Neural Architecture Optimizationcited by 167
  6. 2019Neural Predictor for Neural Architecture Searchcited by 118
  7. 2020BRP-NAS: Prediction-based NAS using GCNscited by 111
  8. 2019NGBoost: Natural Gradient Boosting for Probabilistic Predictioncited by 94
  9. 2021How Powerful are Performance Predictors in Neural Architecture Search?cited by 90
  10. 2021HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmarkcited by 84
  11. 2021Zen-NAS: A Zero-Shot NAS for High-Performance Deep Image Recognitioncited by 73
  12. 2016Understanding Probabilistic Sparse Gaussian Process Approximationscited by 68
  13. 2020Surrogate NAS Benchmarks: Going Beyond the Limited Search Spaces of Tabular NAS Benchmarkscited by 67
  14. 2021TransNAS-Bench-101: Improving Transferability and Generalizability of Cross-Task Neural Architecture Searchcited by 58
  15. 2020Evaluating Efficient Performance Estimators of Neural Architecturescited by 55
  16. 2019Best Practices for Scientific Research on Neural Architecture Searchcited by 54
  17. 2019Understanding Architectures Learnt by Cell-based Neural Architecture Searchcited by 48
  18. 2020Semi-Supervised Neural Architecture Searchcited by 46
  19. 2020A Study on Encodings for Neural Architecture Searchcited by 38
  20. 2022NAS-Bench-Suite-Zero: Accelerating Research on Zero Cost Proxiescited by 36
  21. 2019BlockSwap: Fisher-guided Block Substitution for Network Compression on a Budgetcited by 26
  22. 2021EPE-NAS: Efficient Performance Estimation Without Training for Neural Architecture Searchcited by 24
  23. 2019Deep Neural Architecture Search with Deep Graph Bayesian Optimizationcited by 18
  24. 2023Framework and Benchmarks for Combinatorial and Mixed-variable Bayesian Optimizationcited by 16
  25. 2023Neural Architecture Design and Robustness: A Datasetcited by 15
  26. 2023Multi-Predict: Few Shot Predictors For Efficient Neural Architecture Searchcited by 6
  27. 2023Are Neural Architecture Search Benchmarks Well Designed? A Deeper Look Into Operation Importancecited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 7; 2026: 2.

11 papers.

  1. 2025Regression Language Models for Codecited by 11
  2. 2024Scalable Reinforcement Learning-based Neural Architecture Searchcited by 5
  3. 2025Transferrable Surrogates in Expressive Neural Architecture Search Spacescited by 5
  4. 2025Training-free Neural Architecture Search through Variance of Knowledge of Deep Network Weightscited by 3
  5. 2024Designing a Classifier for Active Fire Detection from Multispectral Satellite Imagery Using Neural Architecture Searchcited by 2
  6. 2025AnalogNAS-Bench: A NAS Benchmark for Analog In-Memory Computingcited by 1
  7. 2025Match & Choose: Model Selection Framework for Fine-tuning Text-to-Image Diffusion Modelscited by 1
  8. 2025ONNX-Net: Towards Universal Representations and Instant Performance Prediction for Neural Architecturescited by 1
  9. 2025Universal Neural Architecture Space: Covering ConvNets, Transformers and Everything in Betweencited by 1
  10. 2026STLGT: A Scalable Trace-Based Linear Graph Transformer for Tail Latency Prediction in Microservicescited by 0
  11. 2026Prediction-powered Neural Architecture Searchcited by 0