A survey of manifold learning and its applications for multimedia, Hannes Fassold, Proc. ICVSP, 2023

You are in taxonomy page

A survey of manifold learning and its applications for multimedia, Hannes Fassold, Proc. ICVSP, 2023

The survey paper gives an introduction into manifold learning and how it is employed for important application fields (similarity search, image classification, synthesis & enhancement, video analysis, 3D data processing, nonlinear dimension reduction) in multimedia. Furthermore, available open source software frameworks for manifold learning are presented.

Lecture notes on reinforcement learning

Reinforcement learning is an appealing subject. Firstly, it is a very general concept: an agent interacts with an environment with the goal to maximize
the rewards it receives from the environment. The environment is random and provides states and rewards to the agent, while the agent chooses actions according to a possibly random policy. The goal is to find policies that maximise the expected value of all future rewards. Because reinforcement learning is such a general concept, it encompasses many real-world applications of machine learning and artificial intelligence

Tutorial paper on Deep Learning for Graphs

The adaptive processing of graph data is a long-standing research topic that has been lately consolidated as a theme of major interest in the deep learning community. The snap increase in the amount and breadth of related research has come at the price of little systematization of knowledge and attention to earlier literature. This resource is a tutorial introduction to the field of deep learning for graphs. It favors a consistent and progressive presentation of the main concepts and architectural aspects over an exposition of the most recent literature, for which the reader is referred to available surveys. The essay takes a top-down view of the problem, introducing a generalized formulation of graph representation learning based on a local and iterative approach to structured information processing. Moreover, it introduces the basic building blocks that can be combined to design novel and effective neural models for graphs.
Link to the resource: https://arxiv.org/pdf/1912.12693.pdf

The Silent (R)evolution of SAT

The Propositional Satisfiability problem (SAT) was the first to be shown NP-complete by Cook and Levin. SAT remained the embodiment of theoretical worst-case hardness. However, in stark contrast to its theoretical hardness, SAT has emerged as a central target problem for efficiently solving a wide variety of computational problems. SAT solving technology has continuously advanced since a breakthrough around the millennium, which catapulted practical SAT solving ahead by orders of magnitudes. Today, the many flavours of SAT technology can be found in all areas of technological innovation.

Algorithms for manifold learning

The technical report presents popular methods for mapping data into a low-dimensional manifold (nonlinear dimensionality reduction). Specifically, it presents Isomap, Locally Linear Embedding, Laplacian Eigenmaps, Semidefinite Embedding and variants of those.

Geometric data analysis based on manifold learning with applications for image understanding

The conference paper gives in the first section a brief and easy understandable introduction into the basics of Riemannian geometry. Furthermore, it gives a review of classical methods for mapping data into a low-dimensional manifold / nonlinear dimensionality reduction like Local Linear Embedding (LLE), Isometric Feature Mapping (ISOMAP) and Local Riemannian Manifold Learning (LRML).