Decision Surfaces. Support Vector Machines
This lecture overviews Decision Surfaces and, in particular, Support Vector Machines that have many applications in Machine Learning and Pattern Recognition. It covers the following topics in detail: Decision surfaces. Hyperplanes. Non-linear Decision Surfaces. Quadratic (2nd degree polynomial) surfaces, Hyperellipsoid/Hyperparaboloid. Support Vector Machines, Margin Maximization, Lagrangian Primal/Dual Problem, Kernel SVM.
Should you require access to the resource, please contact the author directly.
Label Propagation
This lecture overviews Label Propagation that has many applications in pattern recognition (semi-supervised learning) and in the study of diffusion processes. It covers the following topics in detail: Graph construction approaches (Adjacency Matrix Construction, Graph Weighting, Simultaneous Graph Construction and Weighting). Label Inference Methods (Graph Min-cut, Markov Random Fields, Gaussian Random Fields, Local and Global Consistency, Label Propagation on Data with Multiple Representations, Label Propagation on Hypergraphs). Label Propagation for Deep Learning.
Data Clustering
This lecture overviews Data Clustering that has many applications in e.g., facial image clustering, signal/image clustering, concept creation. It covers the following topics in detail: Clustering Definitions. Distance measures, Mahalanobis distance, Euclidean distance, Lp norm, L1 Norm Similarity measures, Cosine similarity, Correlation coefficient. Distance Functions between a Point and a Set. Distance Functions between two Sets. Clustering algorithm categories: Exhaustive Clustering. Sequential Clustering, Maximin algorithm. Clustering by optimization, K-means algorithm, ISODATA algorithm. Fuzzy clustering. Vector Quantization, Voronoi regions, LVQs. Graph-based clustering, N-Cut Graph Clustering, Spectral graph clustering.
Should you require access to the resource, please contact the author directly.
Distance-based Classification
This lecture overviews Distance-based Classification that has many applications in classification. It covers the following topics in detail: k-Nearest neighbor classification, Nearest neighbor graphs Supervised Learning Vector Quantization, LVQ1/2/3.
Should you require access to the resource, please contact the author directly.
Human-Centered AI for Autonomous Vehicles
This lecture overviews human-centric AI methods that can be utilized to facilitate visual interaction between humans and autonomous vehicles (e.g., through gestures captured by RGB cameras), in order to ensure their safe and successful cooperation in real-world scenarios. Such methods should: a) demonstrate increased visual perception accuracy to understand human visual cues, b) be robust to input data variations, in order to successfully handle illumination/background/scale changes that are typically encountered in real-world scenarios, and c) produce timely predictions to ensure safety, which is a critical aspect of autonomous vehicles’ applications.
Should you require access to the resource, please contact the author directly.
Introduction to Machine Learning
This lecture will cover the basic concepts of Machine Learning to alleviate inconsistencies towards concept and notation accuracy. Supervised, self-supervised, unsupervised, semi-supervised learning. Multi-task Machine Learning. Classification, regression. Object detection, Object tracking. Clustering. Dimensionality reduction, data retrieval. Artificial Neural Networks. Adversarial Machine Learning. Generative Machine Learning. Temporal Machine learning (Recurrent Neural Networks). Continual Learning (few-shot learning, online learning). Reinforcement Learning. Adaptive learning (Knowledge Distillation, Domain adaptation, Transfer learning, Activation Pattern Analysis, Federated learning/Collaborative learning, Ensemble learning). Precise mathematical definitions of ML tasks will be presented.
Should you require access to the resource, please contact the author directly.