Methods › General › Activation Functions › Maxout
Maxout
Introduced by Ian J. Goodfellow et al. in Maxout Networks
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
The Maxout Unit is a generalization of the ReLU and the leaky ReLU functions. It is a piecewise linear function that returns the maximum of the inputs, designed to be used in conjunction with dropout. Both ReLU and leaky ReLU are special cases of Maxout.
f(x) = max(wᵀ₁x + b₁, wᵀ₂x + b₂)
The main drawback of Maxout is that it is computationally expensive as it doubles the number of parameters for each neuron.
In Syntology name not resolved (unknown_name). The lookup matched no method, paper or venue by this name; that asserts nothing about whether code exists.
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
Papers archive 2025-07-28
archive papers tagged: 50 · with a code link: 15 · where Syntology ran a sample: 4 (4 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (4 of 50 tagged: 4 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 50 of 50, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
Deep-ICE: The first globally optimal algorithm for empirical risk minimization of two-layer maxout and ReLU networks 9 May 2025 · 0 repositories · arXiv:2505.05740
-
Depth-Bounds for Neural Networks via the Braid Arrangement 13 Feb 2025 · 0 repositories · arXiv:2502.09324
-
Deep Maxout Network-based Feature Fusion and Political Tangent Search Optimizer enabled Transfer Learning for Thalassemia Detection 3 Aug 2023 · 0 repositories · arXiv:2308.02029
-
Query Structure Modeling for Inductive Logical Reasoning Over Knowledge Graphs 23 May 2023 · 1 repository · arXiv:2305.13585
-
Error bounds for maxout neural network approximations of model predictive control 18 Apr 2023 · 0 repositories · arXiv:2304.08779
-
Expected Gradients of Maxout Networks and Consequences to Parameter Initialization 17 Jan 2023 · 1 repository · arXiv:2301.06956
-
Deep Maxout Network Gaussian Process 8 Aug 2022 · 0 repositories · arXiv:2208.04468
-
On the Number of Regions of Piecewise Linear Neural Networks 17 Jun 2022 · 0 repositories · arXiv:2206.08615
-
Learning with Stochastic Orders 27 May 2022 · 1 repository · arXiv:2205.13684
-
Smooth Maximum Unit: Smooth Activation Function for Deep Networks Using Smoothing Maximum Technique 1 Jan 2022 · 0 repositories
-
Learning Discriminative Shrinkage Deep Networks for Image Deconvolution 27 Nov 2021 · 1 repository · arXiv:2111.13876
-
Time-Frequency Localization Using Deep Convolutional Maxout Neural Network in Persian Speech Recognition 9 Aug 2021 · 0 repositories · arXiv:2108.03818
-
On the Expected Complexity of Maxout Networks 1 Jul 2021 · 1 repository · arXiv:2107.00379Syntology official (archive's flag): 3 ran · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified (of 4 harvested samples)
-
Revisiting 2D Convolutional Neural Networks for Graph-based Applications 23 May 2021 · 0 repositories · arXiv:2105.11016
-
Maximum and Leaky Maximum Propagation 21 May 2021 · 0 repositories · arXiv:2105.10277
-
Performance Evaluation of Deep Convolutional Maxout Neural Network in Speech Recognition 4 May 2021 · 0 repositories · arXiv:2105.01399
-
Sharp bounds for the number of regions of maxout networks and vertices of Minkowski sums 16 Apr 2021 · 0 repositories · arXiv:2104.08135
-
A new semi-supervised self-training method for lung cancer prediction 17 Dec 2020 · 0 repositories · arXiv:2012.09472
-
Activate or Not: Learning Customized Activation 10 Sep 2020 · 5 repositories · arXiv:2009.04759
-
Deep Neural-Kernel Machines 13 Jul 2020 · 0 repositories · arXiv:2007.06655
-
CSPNet: A New Backbone that can Enhance Learning Capability of CNN 27 Nov 2019 · 123 repositories · arXiv:1911.11929
-
Best Practices for Convolutional Neural Networks Applied to Object Recognition in Images 29 Oct 2019 · 0 repositories · arXiv:1910.13029
-
Graph-Preserving Grid Layout: A Simple Graph Drawing Method for Graph Classification using CNNs 26 Sep 2019 · 0 repositories · arXiv:1909.12383
-
Towards Learning Affine-Invariant Representations via Data-Efficient CNNs 31 Aug 2019 · 0 repositories · arXiv:1909.00114
-
Conditional Computation for Continual Learning 16 Jun 2019 · 0 repositories · arXiv:1906.06635
-
Max-plus Operators Applied to Filter Selection and Model Pruning in Neural Networks 19 Mar 2019 · 1 repository · arXiv:1903.08072
-
Paragraph-level Neural Question Generation with Maxout Pointer and Gated Self-attention Networks 1 Oct 2018 · 1 repository
-
Competition vs. Concatenation in Skip Connections of Fully Convolutional Networks 20 Jul 2018 · 0 repositories · arXiv:1807.07803
-
A Tropical Approach to Neural Networks with Piecewise Linear Activations 22 May 2018 · 0 repositories · arXiv:1805.08749
-
Single Image Super-Resolution Using Lightweight CNN with Maxout Units 7 Nov 2017 · 0 repositories · arXiv:1711.02321
-
Neuron Pruning for Compressing Deep Networks using Maxout Architectures 21 Jul 2017 · 0 repositories · arXiv:1707.06838
-
Adversarially Learned Inference 2 Jun 2016 · 9 repositories · arXiv:1606.00704Syntology official (archive's flag): 1 ran · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified (of 4 harvested samples) · 2 pointer-only (licence)
-
The IBM 2016 English Conversational Telephone Speech Recognition System 27 Apr 2016 · 0 repositories · arXiv:1604.08242
-
DehazeNet: An End-to-End System for Single Image Haze Removal 28 Jan 2016 · 3 repositories · arXiv:1601.07661
-
A Unified Gradient Regularization Family for Adversarial Examples 19 Nov 2015 · 0 repositories · arXiv:1511.06385
-
Manifold Regularized Deep Neural Networks using Adversarial Examples 19 Nov 2015 · 0 repositories · arXiv:1511.06381
-
Competitive Multi-scale Convolution 18 Nov 2015 · 0 repositories · arXiv:1511.05635
-
A Light CNN for Deep Face Representation with Noisy Labels 9 Nov 2015 · 19 repositories · arXiv:1511.02683Syntology official (archive's flag): 1 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified (of 2 harvested samples) · 1 pointer-only (licence)
-
Batch-normalized Maxout Network in Network 9 Nov 2015 · 1 repository · arXiv:1511.02583Syntology 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified (of 3 harvested samples) · 3 pointer-only (licence)
-
On the Importance of Normalisation Layers in Deep Learning with Piecewise Linear Activation Units 3 Aug 2015 · 0 repositories · arXiv:1508.00330
-
Random Maxout Features 11 Jun 2015 · 0 repositories · arXiv:1506.03705
-
The IBM 2015 English Conversational Telephone Speech Recognition System 21 May 2015 · 0 repositories · arXiv:1505.05899
-
Improved Microaneurysm Detection using Deep Neural Networks 17 May 2015 · 0 repositories · arXiv:1505.04424
-
Training deep neural networks with low precision multiplications 22 Dec 2014 · 1 repository · arXiv:1412.7024
-
On the Number of Linear Regions of Deep Neural Networks 8 Feb 2014 · 0 repositories · arXiv:1402.1869
-
Improving Deep Neural Networks with Probabilistic Maxout Units 20 Dec 2013 · 0 repositories · arXiv:1312.6116
-
From Maxout to Channel-Out: Encoding Information on Sparse Pathways 18 Nov 2013 · 0 repositories · arXiv:1312.1909
-
Learned-Norm Pooling for Deep Feedforward and Recurrent Neural Networks 7 Nov 2013 · 0 repositories · arXiv:1311.1780
-
End-to-End Text Recognition with Hybrid HMM Maxout Models 7 Oct 2013 · 0 repositories · arXiv:1310.1811
-
Maxout Networks 18 Feb 2013 · 7 repositories · arXiv:1302.4389Syntology 0 ran · 3 unverified (of 3 harvested samples)
Tasks archive 2025-07-28
20 shown of 52 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections