Foundational Papers

These are the backbone papers behind modern deep learning. This page is the map: each focused database keeps the paper list compact, link-first, and easy to extend.

Focused Databases

DatabaseScope
Learning and BackpropagationBackpropagation, early CNNs, LSTMs, deep belief nets, and early deep representation learning.
Architecture and TrainabilityHighway networks, ResNets, skip connections, pre-activation residual units, dense connections, Transformers, and neural ODEs.
Initialization, Activations, and NormalizationXavier, ReLU, He initialization, BatchNorm, LayerNorm, GroupNorm, RMSNorm, and modern activations.
Optimization and RegularizationAdam, AdamW, dropout, gradient clipping, label smoothing, mixup, learning-rate schedules, and sharpness-aware training.
Generalization and ScalingGeneralization puzzles, memorization, lottery tickets, NTK, double descent, scaling laws, and compute-optimal training.

Milestone Map

StageKey PapersPrimary Database
Learning foundationsBackpropagation, early CNNs, LSTM, deep belief nets, deep autoencodersLearning and Backpropagation
Trainable depthHighway Networks, ResNet, Identity Mappings, Stochastic Depth, DenseNet, Transformer, Neural ODEsArchitecture and Trainability
Stable signal flowXavier, ReLU, He initialization, GELU, BatchNorm, LayerNorm, GroupNorm, RMSNormInitialization, Activations, and Normalization
Training recipesGradient clipping, Adam, Dropout, label smoothing, AdamW, mixup, AugMix, SAM optimizerOptimization and Regularization
Generalization and scaleRethinking Generalization, Lottery Ticket, NTK, Double Descent, Scaling Laws, ChinchillaGeneralization and Scaling

Suggested Paths

PathRead
Training mechanicsBackpropagation Xavier ReLU He initialization BatchNorm ResNet LayerNorm.
Deep architecturesHighway Networks ResNet Identity Mappings DenseNet Transformer Neural ODEs.
Practical optimizationGradient clipping Dropout Adam label smoothing AdamW mixup SAM.
Modern scalingRethinking Generalization Lottery Ticket NTK Double Descent Scaling Laws Chinchilla.