ScholarGate
Βοηθός

Σύγκριση μεθόδων

Εξετάστε τις επιλεγμένες μεθόδους δίπλα-δίπλα· οι γραμμές που διαφέρουν επισημαίνονται.

Μέθοδοι Κλίσης Πολιτικής×Βαθιά Ενισχυτική Μάθηση×
ΠεδίοΜηχανική ΜάθησηΒαθιά Μάθηση
ΟικογένειαMachine learningMachine learning
Έτος προέλευσης19922015
ΔημιουργόςRonald Williams (REINFORCE); Sutton et al. (policy gradient theorem)Mnih, V. et al. (DQN)
ΤύποςPolicy-based reinforcement learningSequential decision-making (agent–environment interaction)
Θεμελιώδης πηγήWilliams, R. J. (1992). Simple statistical gradient-following algorithms for connectionist reinforcement learning. Machine Learning, 8(3–4), 229–256. DOI ↗Mnih, V. et al. (2015). Human-Level Control through Deep Reinforcement Learning. Nature, 518, 529–533. DOI ↗
Εναλλακτικές ονομασίεςREINFORCE, actor-critic, policy optimization, politika gradyanıDerin Pekiştirmeli Öğrenme (DQN / PPO / A3C), derin pekiştirmeli öğrenme, deep RL, DRL
Συναφείς44
ΣύνοψηPolicy gradient methods are reinforcement-learning algorithms that optimize a parameterized policy directly by gradient ascent on the expected return, rather than learning action-values and acting greedily. Founded on Ronald Williams' 1992 REINFORCE algorithm and the policy gradient theorem of Sutton and colleagues (2000), they naturally handle stochastic and continuous action spaces and underpin modern actor-critic and deep-RL algorithms.Deep Reinforcement Learning combines neural networks with reinforcement learning so an agent learns by interacting with an environment, popularised by Mnih and colleagues' 2015 Nature work on human-level Atari control. Instead of learning from a fixed labelled dataset, the agent takes actions, observes rewards, and gradually shapes a policy that maximises long-run return.
ScholarGateΣύνολο δεδομένων
  1. v1
  2. 2 Πηγές
  3. PUBLISHED
  1. v1
  2. 2 Πηγές
  3. PUBLISHED

Μετάβαση στην αναζήτηση Λήψη διαφανειών

ScholarGateΣύγκριση μεθόδων: Policy Gradient · Deep Reinforcement Learning. Ανακτήθηκε στις 2026-06-17 από https://scholargate.app/el/compare