mlsanityData to 14 Oct 2026

Maximum and Leaky Maximum Propagation

arXiv 2105.10277 ↗First posted 21 May 2021cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

In this work, we present an alternative to conventional residual connections, which is inspired by maxout nets. This means that instead of the addition in residual connections, our approach only propagates the maximum value or, in the leaky formulation, propagates a percentage of both. In our evaluation, we show on different public data sets that the presented approaches are comparable to the residual connections and have other interesting properties, such as better generalization with a constant batch normalization, faster learning, and also the possibility to generalize without additional activation functions. In addition, the proposed approaches work very well if ensembles together with residual networks are formed. https://atreus.informatik.uni-tuebingen.de/seafile/d/8e2ab8c3fdd444e1a135/?p=%2FMaximumPropagation&mode=list

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 33 of its 84 references that are in the map. First posted on arXiv 21 May 2021.

Builds on

33 of 84 references.

References · 33

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  4. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,349
  5. 2014Going Deeper with Convolutionscited by 4,241
  6. 2013Network In Networkcited by 605
  7. 2020Deep Learning for Person Re-identification: A Survey and Outlookcited by 549
  8. 2017Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Modelscited by 206
  9. 2013Maxout Networkscited by 187
  10. 2017Computer Vision for Autonomous Vehicles: Problems, Datasets and State of the Artcited by 182
  11. 2019Knowledge-aware Graph Neural Networks with Label Smoothness Regularization for Recommender Systemscited by 107
  12. 2019On the Impact of the Activation Function on Deep Neural Networks Trainingcited by 102
21 more
  1. 2020Revisiting Batch Normalization for Improving Corruption Robustnesscited by 55
  2. 2019Unsupervised Representation Learning for Gaze Estimationcited by 42
  3. 2016The AGI Containment Problemcited by 36
  4. 2021TEyeD: Over 20 million real-world eye images with Pupil, Eyelid, and Iris 2D and 3D Segmentations, 2D and 3D Landmarks, 3D Eyeball, Gaze Vector, and Eye Movement Typescited by 32
  5. 2016PupilNet: Convolutional Neural Networks for Robust Pupil Detectioncited by 20
  6. 2020Learning Collision-Free Space Detection from Stereo Images: Homography Matrix Brings Better Data Augmentationcited by 19
  7. 2015ElSe: Ellipse Selection for Robust Pupil Detection in Real-World Environmentscited by 16
  8. 2017PupilNet v2.0: Convolutional Neural Networks for CPU based real time Robust Pupil Detectioncited by 15
  9. 2020Fully Convolutional Neural Networks for Raw Eye Tracking Data Segmentation, Generation, and Reconstructioncited by 12
  10. 2020Rotated Ring, Radial and Depth Wise Separable Radial Convolutionscited by 12
  11. 2018Eye movement velocity and gaze data generator for evaluation, robustness testing and assess of eye tracking software and visualization toolscited by 11
  12. 2019Learning to Validate the Quality of Detected Landmarkscited by 11
  13. 2020Weight and Gradient Centralization in Deep Neural Networkscited by 11
  14. 2017Fast camera focus estimation for gaze-based focus controlcited by 10
  15. 2018Eye movement simulation and detector creation to reduce laborious parameter adjustmentscited by 10
  16. 2019Training Decision Trees as Replacement for Convolution Layerscited by 10
  17. 2021The Gaze and Mouse Signal as additional Source for User Fingerprints in Browser Applicationscited by 10
  18. 2020Multi Layer Neural Networks as Replacement for Pooling Operationscited by 8
  19. 2020Explainable Online Validation of Machine Learning Models for Practical Applicationscited by 8
  20. 2021A Multimodal Eye Movement Dataset and a Multimodal Eye Movement Segmentation Analysiscited by 7
  21. 20211000 Pupil Segmentations in a Second using Haar Like Features and Statistical Learningcited by 7

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 3.

4 papers.

  1. 2022Pistol: Pupil Invisible Supportive Tool to extract Pupil, Iris, Eye Opening, Eye Movements, Pupil and Iris Gaze Vector, and 2D as well as 3D Gazecited by 9
  2. 2021Tensor Normalization and Full Distribution Trainingcited by 3
  3. 2022GroupGazer: A Tool to Compute the Gaze per Participant in Groups with integrated Calibration to Map the Gaze Online to a Screen or Beamer Projectioncited by 2
  4. 2022Technical Report: Combining knowledge from Transfer Learning during training and Wide Resnetscited by 0