AI Atlas › Uczenie ze wzmocnieniem i decyzje Podstawy RL Reinforcement learning fundamentals · id rlbasics
Stany, akcje, nagrody i strategie działania.
States, actions, rewards and policies.
In one sentence To opis uczenia przez działanie: sytuacja, możliwe ruchy, ich oceny i sposób wybierania kolejnych kroków.
This describes learning through action: the situation, available moves, their scores and how to choose the next steps.
Problem it solves Zanim program zacznie się uczyć, musisz określić, co widzi, co może zrobić i po czym rozpoznać dobry wynik.
Before software can learn, you must define what it observes, what it can do and how to recognise a good outcome.
What it does Zapisuje sytuacje, działania i oceny skutków, łącząc bieżący wybór z oczekiwaną korzyścią z dalszych kroków.
It records situations, actions and outcome scores, connecting a current choice with expected benefits from later steps.
Everyday example W grze określasz pozycję postaci, dozwolone ruchy i punktowanie dotarcia do celu, żeby program miał z czego się uczyć.
In a game, you define the character's position, allowed moves and scoring for reaching the goal so software has a learning setup.
This concerns you if you… chcesz opisać zadanie uczenia przez próby musisz określić działania, obserwacje i ocenę wyniku want to describe a trial based learning task need to define actions, observations and outcome scores Examples MDP · policy · value function Subtopics Częściowa obserwowalność (POMDP) (Partially observable MDPs (POMDP)) : Decision-making when observations do not reveal the full state of the world, so the agent must combine information over time, e.g. through a belief state or network memory.Funkcje wartości i równanie Bellmana (Value functions and the Bellman equation) : How much reward to expect from a state or action; the Bellman equation links the value now to the value one step later. The basis of Q-learning and actor–critic methods.Proces decyzyjny Markowa (MDP) (Markov decision processes (MDP)) : A formal model of sequential decision-making in RL: states, actions, transition probabilities, rewards and a discount factor. One of the standard starting points for describing RL algorithms.Programowanie dynamiczne: iteracja wartości i polityki (Dynamic programming for MDPs (value and policy iteration)) : Computing an optimal policy when the transition and reward model is fully known, by repeatedly applying the Bellman equations. The reference point for methods that learn from trials.Projektowanie i kształtowanie nagrody (Reward design and reward shaping) : Specifying the reward signal so the agent learns the intended behaviour, and adding auxiliary rewards that speed up learning without changing the goal.Uczenie z różnic czasowych (TD) i metody Monte Carlo (Temporal-difference learning and Monte Carlo methods) : Learning value functions from experience without an environment model: Monte Carlo averages full episode returns, while TD updates estimates after every step using the next state's estimate.Wielorękie bandyty: eksploracja kontra eksploatacja (Multi-armed and contextual bandits) : Repeatedly choosing among options with unknown rewards while balancing exploration of new possibilities against exploitation of the current best. Used in recommendation and adaptive online experimentation.Środowiska i symulatory do RL (RL environments and simulators) : The standard reset/step interface and ready-made environments where agents practise; also physics simulators for robots.„Śmiertelna triada” w RL (The deadly triad in reinforcement learning) : Generalising value functions with approximators, and the instability arising from combining approximation, bootstrapping and off-policy learning.Sources