mlsanityData to 7 Oct 2026

Faster gaze prediction with dense networks and Fisher pruning

arXiv 1801.05787 ↗First posted 17 Jan 2018cs.CV stat.MLcited by 92

Not on the map. Why?

Abstract (arXiv)

Predicting human fixations from images has recently seen large improvements by leveraging deep representations which were pretrained for object recognition. However, as we show in this paper, these networks are highly overparameterized for the task of fixation prediction. We first present a simple yet principled greedy pruning method which we call Fisher pruning. Through a combination of knowledge distillation and Fisher pruning, we obtain much more runtime-efficient architectures for saliency prediction, achieving a 10x speedup for the same AUC performance as a state of the art network on the CAT2000 dataset. Speeding up single-image gaze prediction is important for many real-world applications, but it is also a crucial step in the development of video saliency models, where the amount of data to be processed is substantially larger.

Key facts. Cited by 92 papers in the map (data through 7 Oct 2026). First posted on arXiv 17 Jan 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 22; 2022: 15; 2023: 18; 2024: 15; 2025: 16; 2026: 6.

92 papers.

Sort:

By area:

  1. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 464
  2. 2021Training Neural Networks with Fixed Sparse Maskscited by 212
  3. 2021Zero-Cost Proxies for Lightweight NAScited by 194
  4. 2021Fishr: Invariant Gradient Variances for Out-of-Distribution Generalizationcited by 186
  5. 2022A Fast Post-Training Pruning Framework for Transformerscited by 161
  6. 2023Neural Architecture Search: Insights from 1000 Paperscited by 136
  7. 2021Group Fisher Pruning for Practical Network Compressioncited by 130
  8. 2021How Powerful are Performance Predictors in Neural Architecture Search?cited by 88
  9. 2022Make Sharpness-Aware Minimization Stronger: A Sparsified Perturbation Approachcited by 70
  10. 2022A Comprehensive Survey on Model Quantization for Deep Neural Networks in Image Classificationcited by 64
  11. 2023Parameter-Efficient Fine-Tuning without Introducing New Latencycited by 59
  12. 2021M-FAC: Efficient Matrix-Free Approximations of Second-Order Informationcited by 54
38 more
  1. 2021Differentiable Subset Pruning of Transformer Headscited by 46
  2. 2021MQBench: Towards Reproducible and Deployable Model Quantization Benchmarkcited by 38
  3. 2021Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safetycited by 35
  4. 2021SOSP: Efficiently Capturing Global Correlations by Second-Order Structured Pruningcited by 30
  5. 2021DECORE: Deep Compression with Reinforcement Learningcited by 29
  6. 2025GuidedQuant: Large Language Model Quantization via Exploiting End Loss Guidancecited by 28
  7. 2023HomoDistil: Homotopic Task-Agnostic Distillation of Pre-trained Transformerscited by 25
  8. 2021Image Cropping on Twitter: Fairness Metrics, their Limitations, and the Importance of Representation, Design, and Agencycited by 22
  9. 2023EMQ: Evolving Training-free Proxies for Automated Mixed Precision Quantizationcited by 22
  10. 2023Zero-Shot Neural Architecture Search: Challenges, Solutions, and Opportunitiescited by 21
  11. 2023Unified Low-Resource Sequence Labeling by Sample-Aware Dynamic Sparse Finetuningcited by 21
  12. 2021ProxyBO: Accelerating Neural Architecture Search via Bayesian Optimization with Zero-cost Proxiescited by 20
  13. 2021GradSign: Model Performance Inference with Theoretical Insightscited by 18
  14. 2022Fine-Tuning Pre-Trained Language Models Effectively by Optimizing Subnetworks Adaptivelycited by 18
  15. 2023TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edgecited by 18
  16. 2023Why is the State of Neural Network Pruning so Confusing? On the Fairness, Comparison Setup, and Trainability in Network Pruningcited by 16
  17. 2021Differentiable Network Pruning for Microcontrollerscited by 15
  18. 2022No Parameters Left Behind: Sensitivity Guided Adaptive Learning Rate for Training Large Transformer Modelscited by 15
  19. 2023Patch-wise Mixed-Precision Quantization of Vision Transformercited by 15
  20. 2022LiteTransformerSearch: Training-free Neural Architecture Search for Efficient Language Modelscited by 14
  21. 2023Auto-Prox: Training-Free Vision Transformer Architecture Search via Automatic Proxy Discoverycited by 13
  22. 2025Fishers for Free? Approximating the Fisher Information Matrix by Recycling the Squared Gradient Accumulatorcited by 13
  23. 2022Towards Theoretically Inspired Neural Initialization Optimizationcited by 11
  24. 2021Zero-Cost Operation Scoring in Differentiable Architecture Searchcited by 9
  25. 2021Neural Architecture Search as Program Transformation Explorationcited by 8
  26. 2024PALM: Pushing Adaptive Learning Rate Mechanisms for Continual Test-Time Adaptationcited by 8
  27. 2024Is Oracle Pruning the True Oracle?cited by 7
  28. 2022Cyclical Pruning for Sparse Neural Networkscited by 6
  29. 2022Converting Artificial Neural Networks to Spiking Neural Networks via Parameter Calibrationcited by 6
  30. 2024CE-NAS: An End-to-End Carbon-Efficient Neural Architecture Search Frameworkcited by 5
  31. 2025L-SWAG: Layer-Sample Wise Activation with Gradients information for Zero-Shot NAS on Vision Transformerscited by 5
  32. 2021NullaNet Tiny: Ultra-low-latency DNN Inference Through Fixed-function Combinational Logiccited by 4
  33. 2022Automatic Sparse Connectivity Learning for Neural Networkscited by 4
  34. 2023Sparsified Model Zoo Twins: Investigating Populations of Sparsified Neural Network Modelscited by 4
  35. 2023Systematic Investigation of Sparse Perturbed Sharpness-Aware Minimization Optimizercited by 4
  36. 2024Decoupled Weight Decay for Any p Normcited by 4
  37. 2024Efficient Model Compression Techniques with FishLegcited by 4
  38. 2022Adaptive Activation-based Structured Pruningcited by 3

50 most cited of 92 All 92 (JSON)