Scribd Logo
Upload

Welcome to Scribd!

  • 22 views

    Reinforcement Learning

    Uploaded by

    Darshan R Gowda
    This document discusses reinforcement learning and Q-learning. It introduces reinforcement learning as a way for an agent to learn what actions to take through trial and error to maximize rewards without being explicitly told which actions to take. It then defines reinforcement learning as addressing how an autonomous agent can learn optimal actions to achieve goals by receiving rewards or penalties from its environment in response to actions. Finally, it provides an example of Q-learning, an algorithm for reinforcement learning, and notes it converges under certain assumptions like the system being a deterministic Markov decision process.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Reinforcement Learning

    Uploaded by

    Darshan R Gowda
    22 views18 pages
    This document discusses reinforcement learning and Q-learning. It introduces reinforcement learning as a way for an agent to learn what actions to take through trial and error to maximize rewards without being explicitly told which actions to take. It then defines reinforcement learning as addressing how an autonomous agent can learn optimal actions to achieve goals by receiving rewards or penalties from its environment in response to actions. Finally, it provides an example of Q-learning, an algorithm for reinforcement learning, and notes it converges under certain assumptions like the system being a deterministic Markov decision process.

    Original Description:

    Original Title

    Reinforcement_learning.pptx

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses reinforcement learning and Q-learning. It introduces reinforcement learning as a way for an agent to learn what actions to take through trial and error to maximize rewards without being explicitly told which actions to take. It then defines reinforcement learning as addressing how an autonomous agent can learn optimal actions to achieve goals by receiving rewards or penalties from its environment in response to actions. Finally, it provides an example of Q-learning, an algorithm for reinforcement learning, and notes it converges under certain assumptions like the system being a deterministic Markov decision process.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    22 views18 pages

    Reinforcement Learning

    Uploaded by

    Darshan R Gowda
    This document discusses reinforcement learning and Q-learning. It introduces reinforcement learning as a way for an agent to learn what actions to take through trial and error to maximize rewards without being explicitly told which actions to take. It then defines reinforcement learning as addressing how an autonomous agent can learn optimal actions to achieve goals by receiving rewards or penalties from its environment in response to actions. Finally, it provides an example of Q-learning, an algorithm for reinforcement learning, and notes it converges under certain assumptions like the system being a deterministic Markov decision process.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 18

    18CS71 - ARTIFICIAL INTELLIGENCE AND

    MACHINE LEARNING – Module 5


    Part2: Reinforcement Learning
    Module 5: Topics
    ⚫ Instance-Base Learning
    ◦ Introduction
    ◦ k-Nearest Neighbour Learning
    ◦ Locally weighted regression
    ◦ Radial basis function
    ◦ Case-Based reasoning
    ⚫ Reinforcement Learning
    ◦ Introduction
    ◦ The learning task
    ◦ Q-Learning
    Introduction
    ⚫ Reinforcement learning is learning what to do—how to
    map situations to actions—so as to maximize a
    numerical reward signal.
    ⚫ The learner is not told which actions to take, but instead
    must discover which actions yield the most reward by
    trying them.
    ⚫ Neither fully supervised nor completely unsupervised
    Reinforcement – What it is?
    ⚫ Reinforcement learning addresses the question of how an
    autonomous agent that senses and acts in its environment can
    learn to choose optimal actions to achieve its goals. This very
    generic problem covers tasks such as learning to control a mobile
    robot, learning to optimize operations in factories, and learning to
    play board games. Each time the agent performs an action in its
    environment, a trainer may provide a reward or penalty to indicate
    the desirability of the resulting state. For example, when training an
    agent to play a game the trainer might provide a positive reward
    when the game is won, negative reward when it is lost, and zero
    reward in all other states. The task of the agent is to learn from this
    indirect, delayed reward, to choose sequences of actions that
    produce the greatest cumulative reward.
    Example
    Example
    Convergence of Q-learning Algorithm

    Q-Learning algorithm converges under the following assumptions


    • System is a deterministic MDP
    • Immediate reward values are bounded
    • Agent selects actions in such a fashion that it visits every possible state-action
    pair infinitely often

    You might also like