Here is a brief summary of topics discussed, thus far, in classes. Slides and coding scripts will be distributed after each lecture via the Google groups mailing-list. Refer to the home page on how to subscribe.

23/09/2021 -- Introduction to Advanced Machine Learning and basics of digital image processing. The first class introduces Machine Learning, its applications and main research areas, with particular reference to Computer Vision, as well as Deep Learning. Furthermore, it includes: basic terminology of Computer Vision, including image formation, cues for 3D reconstruction, object detection and recognition; linear filtering, including convolutions, smoothing and sharpening.

24/09/2021 -- Basics of smoothing, derivative filters, image representations and performance evaluation. Topics include: filtering for reducing noise, additive i.i.d. noise, box and Gaussian average filters, separable filters, template matching, edge detection, first and second derivative filters; challenges in object recognition, image representation with histograms of colors, distance measures, nearest neighbor recognition; performance evaluation with precision, recall, true positive rate, false positive rate, F1 measure and overall accuracy, ROC and PR curves and the area under the curves; log-loss, brier score and model calibration.

30/09/2021 -- Introduction to image classification. Topics include: image classification, L1 distance between images, k-nearest neighbor classification. review of k-Nearest Neighbor, L1 and L2 distance; hyperparameter tuning; Train and test sets and cross-validation; the curse of dimensionality.

1/10/2021 -- Linear classification. Topics include: linear classification with a parametric approach; Algebraic, visual and geometric interpretations; multi-class SVM ("hinge") loss, softmax and cross-entropy loss, L1, L2 and elastic regularization.

07/10/2021 -- Optimization and Features. Topics include: gradient descent, including the numerical Vs analytical computation of the gradient; numerical gradient for "gradient check"; variants of gradient descent, including stochastic and mini-batch gradient descent; image features and end-to-end training of features.

08/10/2021 -- Backpropagation. Topics include: the computational graph; backpropagation via upstream gradients and local gradients for scalar variables; main patterns in gradient flow; backpropagation with vectors and matrices, by means of the Jacobian matrix and by means of implicit multiplications; introduction of neural networks, including fully connected layers and activation functions.

14/10/2021 -- Neural networks. Topics include: review of backpropagation with vectors and matrices, fully connected layers and activation functions; feature extraction Vs end-to-end learning; interpretation and visualization of intermediate network features.

15/10/2021 -- Pytorch. Topics include: introduction to HW for deep neural networks, including CPUs, GPUs and TPUs; introduction of PyTorch; the fundamental concepts of tensors, autograd and modules, coding examples for defining new modules, adopting nn.functional losses and using data loaders; Pytorch pretrained models.

21/10/2021 -- Introduction to ConvNets. Topics include: dynamic Vs static computation graphs, including applications and frameworks relating to Pytorch; introduction to ConvNets, including applications and the concepts of local patterns and shared parameters and pooling.

22/10/2021 -- ConvNets. Topics include: convolutional layers and the hyperparameters of stride, pad and filter sizes; pooling layers and the hyperparameters of stride and filter sizes.

28/10/2021 -- ConvNets invariance property and visualization. Topics include: ConvNet invariance with respect to translation, scaling and rotation; the importance of depth for DNNs; visualization of ConvNet feature map activations; DNN activation functions.

29/10/2021 -- Training NNs. Topics include: more on activation functions, weight initialization, batch normalization.

04/11/2021 -- NN Optimizers. Topics include: neural network optimizers, learning rate decay, model ensembles.

05/11/2021 -- More on Regularization and Training. In the lecture, there has been further discussion on the regularization techniques of DropOut and BatchNormalization, as well as the introduction of others, including variants of data augmentation, DropConnect, Fractional Max Pool, Stochastic Depth, CutOut and Mixup, hyper-parameter tuning, transfer learning; introduction to semantic segmentation and the difference between things Vs stuff.

18/11/2021 -- Semantic Segmentation. Topics include: (max) unpooling and transpose convolutions; intersection over union; efficient labelling for semantic segmentation; object detection including details on classification and bounding box regression losses, selective search and RCNN.

19/11/2021 -- Object Detection. Topics include: Fast and Faster R-CNN, as well as their building blocks RoI Pool, RoI Align and the Region Proposal Network; one-stage object detectors.

25/11/2021 -- Instance Segmentation. Topics include: instance segmentation with Mask R-CNN; the COCO benchmark; introductory notes on image captioning and 3D detection; introduction of final projects.

26/11/2021 -- Temporal Convolutional Networks and Recurrent Neural Networks. Topics include: introduction to sequence modelling and related computer vision applications, with specific reference to forecasting; Temporal Convolutional Networks (TCN) and the logarithmic interaction distance with convolution with dilation; Recurrent Neural Networks (RNN), including their computational graph, learning and inference.

2/12/2021 -- More on RNNs. Topics include: interpretability of internal state cells; image captioning; brief introduction of visual question answering; gradient flow for RNNs, including exploding and vanishing gradients.

3/12/2021 -- LSTMs and Project Presentations. Topics include: more on gradient flow for RNNs, long short-term memory models (LSTM) and other variants such as GRUs.

9/12/2021 -- Trajectory Forecasting and Transformer Networks. Topics include: the role and modelling of social interaction, head pose and contact in people trajectory forecasting; transformer networks, self-attention including multi-head attention.

10/12/2021 -- Music and Image Transformers and BERT. Topics include: positional encoding, image transformers, relative self-attention and the music transformer; masking, training and fine-tuning BERT; model compression, distillation and teacher-student frameworks.

16/12/2021 -- Object, person and semantic search and meta-learning. Topics include: concluding remarks on relative self-attention and the music transformer; masking, training and fine-tuning BERT; visual object search with local, global and deep representations; triplet loss and siamese networks; re-identification with triplet loss and siamese networks; person search with OIM and QEEPS; semantic search with visual and text embeddings; introductory notes to meta-learning.

18/12/2021 -- Final Project Presentations. Students have delivered their final project presentations.