Computer Engineering is coming soon
We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.
Computer Engineering
Deep Learning
Deep feed-forward networks, CNNs, RNNs and LSTMs, GANs, deep reinforcement learning.
Notes for Deep Learning are being written and reviewed. Until they're up, Aria teaches every topic — ask for any part of the syllabus, with worked numericals.
Coming next in Deep Learning
- Deep learning vs machine learning· notes coming
- History, advantages and challenges of deep learning· notes coming
- Representation learning and deep architectures· notes coming
- Deep learning frameworks: TensorFlow, Keras and PyTorch· notes coming
- Perceptron and feed-forward networks· notes coming
- Forward and backpropagation· notes coming
- Activation functions in deep networks· notes coming
- Loss functions· notes coming
- Hyperparameters: learning rate, momentum and regularisation· notes coming
- Gradient descent and vanishing and exploding gradients· notes coming
- CNN architecture· notes coming
- Convolution, padding and strides· notes coming
- Pooling and fully connected layers· notes coming
- Training a CNN· notes coming
- Recurrent neural networks· notes coming
- Bidirectional and deep RNNs· notes coming
- Encoder–decoder sequence-to-sequence models· notes coming
- Long-term dependencies, LSTM and GRU· notes coming
- Recursive networks and echo state networks· notes coming
- Practical methodology: metrics, baselines and hyperparameters· notes coming
- Deep generative models· notes coming
- Boltzmann machines and deep belief networks· notes coming
- Generative adversarial networks· notes coming
- Types and applications of GANs· notes coming
- Deep reinforcement learning and MDPs· notes coming
- Dynamic programming for reinforcement learning· notes coming
- Q-learning and deep Q-networks· notes coming
- Optimisers: momentum, Nesterov, AdaGrad, RMSProp and Adam· notes coming
- Regularisation: dropout, batch normalisation and early stopping· notes coming
- Autoencoders· notes coming
- Denoising, sparse and contractive autoencoders· notes coming
- LeNet and AlexNet· notes coming
- Backpropagation through time· notes coming
- CNN architectures: VGG, GoogLeNet, ResNet and DenseNet· notes coming
- Object detection and image segmentation· notes coming
- Variational autoencoders· notes coming
- Transfer learning and data augmentation· notes coming
- Explainable and responsible AI· notes coming
- MLOps and model deployment· notes coming
- McCulloch–Pitts neuron and thresholding logic· notes coming
- PCA and singular value decomposition· notes coming
- Weight initialisation and greedy layer-wise pre-training· notes coming
- Learning word representations· notes coming
- Attention mechanism· notes coming
- Evolution of generative AI· notes coming
- Autoregressive models: RNN- and CNN-based· notes coming
- Scaling laws for language models· notes coming
- Structured probabilistic models: directed and undirected graphs· notes coming
- Partition function and d-separation· notes coming
- Sampling from graphical models· notes coming
- Applications of generative AI· notes coming
- Text, image, audio and video generation· notes coming
- Foundation models and large language models· notes coming
- Zero-shot, one-shot and few-shot prompting· notes coming
- Visualising and fooling CNNs: guided backpropagation and Deep Dream· notes coming