• Back to Profile
    •  

     


    I propose to consider the question, 'Can machines think?'
    This should begin with definitions of the meaning of the terms 'machine' and 'think'.
    The definitions might be framed so as to reflect so far as possible the normal use of words,
    but this attitude is dangerous.

    If the meaning of the words 'machine' and 'think' are to be found
    by examining how they are commonly used
    it is difficult to escape the conclusion that the meaning
    and the answer to the question, 'Can machines think?'
    is to be sought in a statistical survey such as a Gallup poll. But this is absurd.
    - A. M. Turing's (1950, p. 433) 'Computing Machinery and Intelligence'

     

     

     

     

    Acting Rationally



    1. The acting rationally approach is concerned with the underlying principles of rational behaviour

    3. The acting rationally approach is supported by:
      1. The Reinforcement Learning-Based Model of Interaction between an Agent and its Environment
      2. The Rationality Assumption or Expected Utility Hypothesis
      3. The Markov Decision Process


    Reinforcement learning-based model
    According to the Reinforcement Learning-Based Model of Interaction between an Agent and its Environment:
    1. An agent is a minded entity that perceives through sensors (e.g. eyes, ears, cameras, etc), acts through effectors (e.g. hands, legs, motors, etc)
    2. An agent also interacts with its environment and learns accordingly

    5. At step t, the agent performs action At and receives observation Ot and the scalar reward Rt
    6. The environment receives action At and emits observation Ot + 1 and the scalar reward Rt + 1

    A rational action is defined as any action that maximizes the expected value of an objective performance measure, given the percept sequence to date — see also the Rationality Assumption or Expected Utility Hypothesis


    Individual decision-making under risk
    1. Decision theory allows for certain computations to be made on the basis of probability-values and utility-values
    2. In addition, the assignment of probability-values and utility-values is constrained by two distinct sets of axioms: the Kolmogorov axioms and the von Neumann-Morgenstern axioms
    See this related entry on decision theory and its pair of axiom sets

    According to the Rationality Assumption or Expected Utility Hypothesis:
    1. If the von Neumann-Morgenstern axioms are satisfied, then the agent is said to be rational and her preferences can be represented by a utility function

    3. Given n alternative courses of action φ1, φ2, …, φn:
    4. Each action φi has a set of associated outcomes (or action-state pairs), where i ∈ ℕ and i ∈ [1, n]
    5. The utility function allows the agent to assign utility-values consistently to each outcome
    6. Choosing the most appropriate course of action φi according to the preference relation ≽ amounts to choosing the course of action with the maximum expected utility


    Andrey Markov
    A Markov Decision Process or MDP is a tuple of the form (S, Φ, P, γ, r), where:
    1. S denotes the set of states (including the initial state s0)
    2. Φ denotes the set of actions available to an agent (viz. φ1, φ2, …)
    3. P denotes the transition probability matrix
    4. γ denotes the discount factor
    5. r denotes the reward function

    7. Reinforcement learning, an instantiation of the acting rationally approach, can be modelled as an MDP
    ∴ The MDP makes the Reinforcement Learning-Based Model of Interaction computationally feasible


    MDP Diagram

    1. Here, the agent begins at the initial state s0
    2. The agent selects φ1 ∈ Φ
    3. There is a non-zero probability P(s1|s0, φ1) that the agent will transition from s0 to a successor state s1 in accordance with the transition probability matrix P
    4.    ⋮
    5. γ denotes the discount factor, where γ ∈ [0, 1)
    6. γ allows for more immediate rewards to count more than later or time-discounted rewards

    Expected total payoff value or cumulative reward = E(r(s0,φ1) + γr(s1,φ2) + γ2r(s2,φ3) + …)


    See also the Rationality Assumption or Expected Utility Hypothesis







    The acting rationally approach
    Reinforcement learning Inverse reinforcement learning

    In reinforcement learning, the reward function r is typically specified upfront

    Conversely, in inverse reinforcement learning, the reward function is not specified upfront. Rather, it must be extracted from observed optimal behaviour


    Given a specified reward function r, the reinforcement learning-based agent attempts to infer an optimal policy π


    Given optimal observed behaviour, the inverse reinforcement learning-based agent attempts to infer the reward function r



    There is growing interest in the acting rationally approach to AI (Russell & Norvig, 1995, 2002, 2010)

     

     

    Background image taken from: https://cdn.asiatatler.com/asiatatler/i/th/2020/02/04101225-aurora-1185464-1920_cover_1920x1280.jpg This website has been coded using html, css, and js and is dedicated to B and H .