Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 56 views

    Artificial Intelligence and Intelligent Agents (F29AI) MDP I: Intro To Markov Decision Processes

    Uploaded by

    Aimee Lemma
    The document discusses Markov Decision Processes (MDPs) which provide a framework for modeling sequential decision making problems under uncertainty. MDPs are defined by states, actions, transition probabilities between states given an action, and reward functions. They allow planning for optimal decisions even when actions have non-deterministic outcomes defined by probabilities. An example of a noisy "grid world" environment is provided where the agent's movement actions do not always produce the intended result. MDPs generalize search problems to stochastic domains and can be solved using techniques like expectimax search or other MDP-specific algorithms.

    Copyright:

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

    Artificial Intelligence and Intelligent Agents (F29AI) MDP I: Intro To Markov Decision Processes

    Uploaded by

    Aimee Lemma
    56 views10 pages
    The document discusses Markov Decision Processes (MDPs) which provide a framework for modeling sequential decision making problems under uncertainty. MDPs are defined by states, actions, transition probabilities between states given an action, and reward functions. They allow planning for optimal decisions even when actions have non-deterministic outcomes defined by probabilities. An example of a noisy "grid world" environment is provided where the agent's movement actions do not always produce the intended result. MDPs generalize search problems to stochastic domains and can be solved using techniques like expectimax search or other MDP-specific algorithms.

    Original Title

    Intro-MDP(1)

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses Markov Decision Processes (MDPs) which provide a framework for modeling sequential decision making problems under uncertainty. MDPs are defined by states, actions, transition probabilities between states given an action, and reward functions. They allow planning for optimal decisions even when actions have non-deterministic outcomes defined by probabilities. An example of a noisy "grid world" environment is provided where the agent's movement actions do not always produce the intended result. MDPs generalize search problems to stochastic domains and can be solved using techniques like expectimax search or other MDP-specific algorithms.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    56 views10 pages

    Artificial Intelligence and Intelligent Agents (F29AI) MDP I: Intro To Markov Decision Processes

    Uploaded by

    Aimee Lemma
    The document discusses Markov Decision Processes (MDPs) which provide a framework for modeling sequential decision making problems under uncertainty. MDPs are defined by states, actions, transition probabilities between states given an action, and reward functions. They allow planning for optimal decisions even when actions have non-deterministic outcomes defined by probabilities. An example of a noisy "grid world" environment is provided where the agent's movement actions do not always produce the intended result. MDPs generalize search problems to stochastic domains and can be solved using techniques like expectimax search or other MDP-specific algorithms.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 10

    Artificial Intelligence and

    Intelligent Agents (F29AI)

    MDP I: Intro to Markov Decision


    Processes
    Arash Eshghi

    Based on slides from Ioannis Konstas @HWU, Verena Rieser @HWU, Dan Klein @UC Berkeley
    So far…
    • How to formalise a real-world problem?
    • How to reach a goal?
    • Blind search: DFS, BFS, UCS
    • Informed search: Greedy, A*
    • How to maximise an outcome if there is an adversary?
    • Adversary search: minimax, alpha-beta pruning
    • What to do if this adversary (environment) is
    probabilistic?
    • Expectimax, maximum expected utilities
    Today
    • Non-deterministic worlds
    • Markov Decision Processes (MDPs)
    • Time-limited values

    How can you plan optimally if your


    actions are non-deterministic?
    (you only have a distribution over the
    outcome)
    Today
    Example: Grid World
    • A maze-like problem
    • The agent lives in a grid
    • Walls block the agent’s path
    • Noisy movement: actions do not always go
    as planned
    • 80% of the time, the action North takes the
    agent North (if there is no wall there)
    • 10% of the time, North takes the agent West;
    10% East
    • If there is a wall in the direction the agent would
    have been taken, the agent stays put
    • The agent receives rewards each time step
    • Small “living” reward each step (can be negative)
    • Big rewards come at the end (good or bad)
    • Goal: maximize sum of rewards
    Grid World Actions
    Deterministic Stochastic
    Grid World Grid World
    Markov Decision Processes

    • An MDP is defined by:


    • A set of states s Î S
    • A set of actions a Î A
    • A transition function T(s, a, s’)
    • Prob that action a from s leads to s’
    i.e., P(s’ | s,a)
    • Also called “the model”
    • A reward function R(s, a, s’)
    • Sometimes just R(s) or R(s’)
    • A start state (or distribution)
    • Maybe a terminal state

    • MDPs are a family of non-deterministic search problems


    • One way to solve them is with expectimax search – but we will
    have a new tool soon
    7
    Video of Demo Gridworld
    Manual Intro
    Gridworld Demo
    What is Markov about MDPs?

    • “Markov” generally means that given the present state,


    the future and the past are independent.

    • For Markov decision processes, “Markov” means action


    outcomes depend only on the current state

    Andrey Markov
    (1856-1922)

    You might also like