Deep Learning
Backprop, optimization, normalization, CNNs, sequence models, and an intro to attention. Assumes ML Foundations.
0 / 28 solved
-
NN Fundamentals & Backprop
Forward and backward passes, the chain rule, vanishing and exploding gradients, initialization.
-
Activations & Optimization
Activation functions, adaptive optimizers, learning-rate schedules, momentum, and sharpness-aware minimization.
-
Regularization & Loss Techniques
BatchNorm vs GroupNorm, dropout, weight decay, and loss-level tricks: label smoothing, focal loss, mixup.
-
Architectures (CNN / RNN)
Convolution, pooling, receptive fields, recurrent nets, and residual connections.
-
Attention & Transformers (intro)
Queries, keys, values, scaled dot-product attention, multi-head attention, positional encoding.
Up next
28 of 28 questions here are still unsolved.