Adobe PDF (1.64 MB)
Title Details:
Deep Reinforcement Learning and Genetic Algorithms
Authors: Symeonidis, Panagiotis
Description:
Abstract:
In chapter seven we will present reinforcement learning algorithms (reinforcement learning) -because it reinforces desired behaviours by rewarding them- where the system recommendation system tries to learn through its interaction with the user. Furthermore, we will analyse recommendation algorithms (Markov Chains, Markov Decision Processes etc.) based on data that evolve over time, i.e. that have a sequence and are in serial order. We will also study Advantage Actor-Critic learning algorithms (A2C) and Deep Q-learning Network (DQN), which are reinforcement learning algorithms that combine its methods with multi-level neural networks. Finally, we will describe the recommendation systems based on genetic algorithms, which simulate the natural phenomenon of evolution and natural selection: the search for the appropriate neighbourhood of the user under consideration is thus initiated by a number of random neighbouring users on the basis of a set of initial assumptions. On this initial population, and if its members are evaluated by means of a fitness function, the new generation of neighbours is generated by reproduction procedures (e.g. crossover, mutation etc.).
Linguistic Editors: Sakellarios, Michalis
Technical Editors: Karatzidis, Dimitrios
Graphic Editors: Symeonidis, Panagiotis
Type: Chapter
Creation Date: 29-05-2023
Item Details:
License: Attribution - NonCommercial - ShareAlike 4.0 International (CC BY-NC-SA 4.0)
Handle http://hdl.handle.net/11419/9579
Bibliographic Reference: Symeonidis, P. (2023). Deep Reinforcement Learning and Genetic Algorithms [Chapter]. In Symeonidis, P. 2023. Intelligent Recommender Systems [Postgraduate textbook]. Kallipos, Open Academic Editions. https://hdl.handle.net/11419/9579
Language: Greek
Is Part of: Intelligent Recommender Systems
Publication Origin: Kallipos, Open Academic Editions