Methods › General
General
Collections are ordered by tagged papers; each shows its 5 most-tagged methods.
Output Functions
10 methods · 38,029 papersSoftmax
37,443 papers
Heatmap
437 papers
Mixture of Logistic Distributions
170 papers
Adaptive Softmax
72 papers
Normalization
38 methods · 32,100 papersLayer Normalization
24,980 papers
Batch Normalization
6,287 papers
Instance Normalization
572 papers
Adaptive Instance Normalization
429 papers
Local Response Normalization
421 papers
Attention Mechanisms
62 methods · 32,081 papersGlobal-Local Attention
56 papers
Location-based Attention
37 papers
Feedforward Networks
15 methods · 31,202 papersDense Connections
29,230 papers
Linear Layer
25,421 papers
Position-Wise Feed-Forward Layer
13,895 papers
Feedforward Network
1,339 papers
Adapter
1,147 papers
Skip Connections
4 methods · 31,178 papersResidual Connection
28,401 papers
Concatenated Skip Connection
3,337 papers
Zero-padded Shortcut Connection
8 papers
Deactivable Skip Connection
1 paper
Regularization
58 methods · 30,804 papersDropout
27,472 papers
Label Smoothing
14,327 papers
Attention Dropout
10,892 papers
Weight Decay
10,713 papers
Entropy Regularization
1,128 papers
Stochastic Optimization
65 methods · 29,207 papersAttention Modules
33 methods · 25,082 papersMulti-Head Attention
24,855 papers
Graph Self-Attention
64 papers
Deformable Attention Module
42 papers
Activation Functions
69 methods · 17,139 papersReLU
10,350 papers
Sigmoid Activation
7,112 papers
Tanh Activation
6,333 papers
Gated Linear Unit
798 papers
Position Embeddings
6 methods · 14,010 papersAbsolute Position Encodings
13,942 papers
Relative Position Encodings
45 papers
Conditional Positional Encoding
7 papers
Neural Tangent Transfer
1 paper
Active Learning
5 methods · 12,227 papersTest
6,434 papers
Learning Rate Schedules
13 methods · 11,277 papersLinear Warmup With Linear Decay
7,076 papers
Cosine Annealing
3,965 papers
Linear Warmup With Cosine Annealing
3,797 papers
Inverse Square Root Schedule
702 papers
Exponential Decay
110 papers
Non-Parametric Regression
9 methods · 4,530 papersGaussian Process
2,473 papers
Energy Based Process
2 papers
Non-Parametric Classification
6 methods · 4,481 papersGaussian Process
2,473 papers
Skip Connection Blocks
42 methods · 4,219 papersResidual Block
2,807 papers
Bottleneck Residual Block
2,049 papers
Inverted Residual Block
823 papers
Dense Block
497 papers
Non-Local Block
168 papers
Knowledge Distillation
11 methods · 4,205 papersModel Compression
5 methods · 4,096 papersPruning
3,874 papers
CORAD
CORAD: Correlation-Aware Compression of Massive Time Series using Sparse Dictionary Coding
1 paper
Generalized Linear Models
3 methods · 3,575 papersDimensionality Reduction
8 methods · 3,470 papersVariational Inference
846 papers
Latent Diffusion Model
366 papers
Initialization
7 methods · 3,032 papersKaiming Initialization
2,931 papers
Xavier Initialization
121 papers
Fixup Initialization
2 papers
SkipInit
2 papers
T-Fixup
2 papers
Loss Functions
43 methods · 3,015 papersFocal Loss
461 papers
Cycle Consistency Loss
448 papers
Triplet Loss
424 papers
GAN Least Squares Loss
421 papers
InfoNCE
418 papers
Self-Supervised Learning
47 methods · 3,010 papersOptimization
34 methods · 2,251 papersGradient Clipping
167 papers
Sharpness-Aware Minimization
142 papers
Fine-Tuning
9 methods · 2,003 papersDiscriminative Fine-Tuning
1,990 papers
Virtual Data Augmentation
4 papers
Leverage Learning
3 papers
Child-Tuning
1 paper
AutoML
10 methods · 1,876 papersFeature Selection
1,602 papers
VEGA
5 papers
Clustering
19 methods · 1,531 papersk-Means Clustering
772 papers
Spectral Clustering
517 papers
Coresets
103 papers
Ensemble Clustering
19 papers
Interpretability
17 methods · 1,371 papersMonte Carlo Dropout
202 papers
Visual Analytics
78 papers
Large Batch Optimization
8 methods · 1,257 papersDeep Tabular Learning
27 methods · 1,237 papersTabNet
29 papers
Distribution Approximation
2 methods · 774 papersNormalizing Flows
771 papers
Discriminators
6 methods · 674 papersPatchGAN
516 papers
Projection Discriminator
135 papers
Window-based Discriminator
14 papers
Multiple Random Window Discriminator
2 papers
Miscellaneous Components
9 methods · 653 papersAuxiliary Classifier
395 papers
Highway Layer
108 papers
Channel Shuffle
80 papers
PixelShuffle
45 papers
Scatter Connection
10 papers
Structured Prediction
4 methods · 473 papersWorking Memory Models
5 methods · 470 papersMemory Network
445 papers
Neural Turing Machine
22 papers
Dynamic Memory Network
19 papers
End-To-End Memory Network
2 papers
Recurrent Entity Network
2 papers
Semi-Supervised Learning Methods
23 methods · 469 papersDistillation
1 method · 415 papersEnsembling
3 methods · 370 papersHyperparameter Search
4 methods · 355 papersRandom Search
256 papers
Population Based Training
18 papers
Twin Networks
1 method · 343 papersSiamese Network
343 papers
Meta-Learning Algorithms
6 methods · 259 papersMeta-augmentation
4 papers
Neural Architecture Search
20 methods · 246 papersNeural Architecture Search
23 papers
ProxylessNAS
17 papers
Lifelong Learning
3 methods · 237 papersSelf-Learning
235 papers
Kernel Methods
2 methods · 234 papersBijective Transformation
1 method · 230 papersAffine Coupling
230 papers
Fourier-related Transforms
4 methods · 225 papersDiscrete Cosine Transform
212 papers
Characteristic Functions
Characteristic Function Estimation for Discrete Probability Distributions
12 papers
Distributed Methods
34 methods · 222 papersLocal SGD
69 papers
Gradient Sparsification
38 papers
Tofu
19 papers
Chimera
16 papers
IMPALA
16 papers
Representation Learning
5 methods · 186 papersPlaystyle Distance
3 papers
Domain Adaptation
18 methods · 164 papersSource Hypothesis Transfer
3 papers
Robustness Methods
5 methods · 157 papersRandomized Smoothing
147 papers
Denoised Smoothing
8 papers
Fishr
2 papers
Randomized Deletion
2 papers
2D Parallel Distributed Methods
2 methods · 150 papersLatent Variable Sampling
3 methods · 149 papersTruncation Trick
136 papers
Latent Optimisation
13 papers
Data Parallel Methods
15 methods · 138 papersLocal SGD
69 papers
Gradient Sparsification
38 papers
ZeRO
9 papers
Accordion
4 papers
ZeRO-Offload
4 papers
Approximate Inference
1 method · 131 papersPrompt Engineering
4 methods · 122 papersKnowPrompt
6 papers
Parameter Norm Penalties
2 methods · 113 papersL1 Regularization
90 papers
Attention
11 methods · 100 papersClass Attention
36 papers
Grouped-query attention
19 papers
Attention Sinks
13 papers
Multi-Query Attention
13 papers
Recommendation Systems
5 methods · 78 papersAdversarial Training
14 methods · 67 papersAdvProp
6 papers
Parameter Sharing
1 method · 63 papersWeight Tying
63 papers
Diffusion Models
3 methods · 60 papersConsistency Models
59 papers
Robust Training
4 methods · 54 papersSelf-adaptive Training
4 papers
Distributions
1 method · 52 papersGumbel Softmax
52 papers
Prioritized Sampling
4 methods · 45 papersIoU-Balanced Sampling
6 papers
Negative Sampling
2 methods · 40 papersSelf-Adversarial Negative Sampling
30 papers
Model Parallel Methods
7 methods · 34 papersAdaptive Activation Functions
13 methods · 29 papersmodReLU
6 papers
Rational Activation function
3 papers
Synchronous Pipeline Parallel
2 methods · 23 papersVariational Optimization
3 methods · 21 papersResidual Normal Distribution
5 papers
Generalized Additive Models
2 methods · 20 papersBase Boosting
2 papers
Cache Replacement Models
1 method · 17 papersParrot
17 papers
Momentum Rules
1 method · 16 papersDemon
16 papers
Markov Chain Monte Carlo
3 methods · 14 papersMetropolis Hastings
10 papers
Sharded Data Parallel Methods
3 methods · 14 papersAir Quality Forecasting
2 methods · 12 papersRule-based systems
3 methods · 12 papersSymbolic rule learning
4 papers
Action Recognition Blocks
2 methods · 8 papersAffinity Functions
4 methods · 8 papersEmbedded Gaussian Affinity
8 papers
Concatenation Affinity
1 paper
Embedded Dot Product Affinity
1 paper
Gaussian Affinity
0 papers
Gated Linear Networks
3 methods · 8 papersInformation Retrieval Methods
3 methods · 8 papersTILDEv2
3 papers
ReInfoSelect
1 paper
Intra-Layer Parallel
2 methods · 8 papersGShard
6 papers
Mesh-TensorFlow
2 papers
Adversarial Attacks
7 methods · 7 papersNetwork Shrinking
2 methods · 7 papersNetAdapt
3 papers
Adaptive Computation
3 methods · 6 papersReplicated Data Parallel
4 methods · 6 papersRobotic Manipulation Models
2 methods · 6 papersSLAM Methods
2 methods · 6 papersDROID-SLAM
6 papers
VDO-SLAM
0 papers
Asynchronous Pipeline Parallel
3 methods · 5 papersFuzzy Logic
1 method · 5 papersGeometric Matching
1 method · 5 papersMesh-Based Simulation Models
1 method · 5 papersMeshGraphNet
5 papers
Out-of-Distribution Example Detection
1 method · 5 papersProbability distribution representation
1 method · 5 papersAugmented Reality Methods
2 methods · 4 papersFeature Matching
3 methods · 4 papersGeneralization
1 method · 4 papersDistributional Generalization
4 papers
Inference Engines
1 method · 4 papersTabular Data Generation
3 methods · 4 papersCTAB-GAN
2 papers
Factorization Machines
2 methods · 3 papersLearning to Rank Models
1 method · 3 papersDCN-V2
3 papers
Playstyle Distance
3 papers
Quantum Methods
4 methods · 3 papersVQSVD
1 paper
Variational Entanglement Detection
0 papers
Rule Learners
1 method · 3 papersAsynchronous Data Parallel
2 methods · 2 papersCrossbow
2 papers
Wavelet Distributed Training
0 papers
Auto Parallel Methods
3 methods · 2 papersConfidence Calibration
1 method · 2 papersInference Attack
2 methods · 2 papersCanvas Method
1 paper
Spectral DeTuning
1 paper
Information Bottleneck
2 methods · 2 papersAggregated Learning
1 paper
ReInfoSelect
1 paper
Label Correction
2 methods · 2 papersLabel Quality Model
1 paper
Long-Range Interaction Layers
1 method · 2 papersLambda Layer
2 papers
Mixture-of-Experts
1 method · 2 papersDSelect-k
2 papers
Self-Training Methods
2 methods · 2 papersself-mem + new data
1 paper
Ternarization
1 method · 2 papersTernary Weight Splitting
2 papers
Binary Neural Networks
1 method · 1 paperBiDet
1 paper
Bot Detection
1 method · 1 paperBIMAN
1 paper
Confidence Estimators
1 method · 1 paperMACEst
1 paper
Control and Decision Systems
2 methods · 1 paperGA-PID/NN-PID
0 papers
Distributed Communication
1 method · 1 paperBlink Communication
1 paper
Hybrid Fuzzing
1 method · 1 paperMEUZZ
1 paper
Hybrid Optimization
2 methods · 1 paperIncident Aggregation Models
1 method · 1 paperGRLIA
1 paper
Manifold Disentangling
1 method · 1 paperPosition Recovery Models
1 method · 1 paperPRNet+
1 paper
Sample Re-Weighting
1 method · 1 paperFast Sample Re-Weighting
1 paper
SpreadsheetCoder
1 paper
Statistical Inference
1 method · 1 paperEstimation Statistics
1 paper
Table Parsing Models
1 method · 1 paperCycle-CenterNet
1 paper
Theorem Proving Models
1 method · 1 paperNeuroTactic
1 paper
Hybrid Parallel Methods
1 method · 0 papersHetPipe
0 papers
Parameter Server Methods
1 method · 0 papersHetPipe
0 papers