Announcements ▪ HW1 grades are now available on Submitty ▪ For HW2, we have made the Overleaf template available, should you prefer a digital workflow (announced on Piazza on Feb 1) ▪ https://www.overleaf.com/read/sxgtybzsvkbx#da6f1d ▪ Even if you type the answers for the submission, we recommend that you first write the answers on a piece of paper to simulate the exam conditions. 1 MDP: Markov Decision Processes CSCI 4150: Introduction to Artificial Intelligence (Spring 2025) Oshani Seneviratne Assistant Professor in Computer Science [email protected] February 04, 2025 Acknowledgment:
Most of the content in these slides was adapted from the slides created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley.
All CS188 materials are available at http://ai.berkeley.edu. Non-Deterministic Search When is non- deterministic search
useful? When you know what
actions are available to
you and what they
might do, you're not
entirely sure what
outcome will occur. You need to compute
policies that take into
account all of the
different outcomes that
might occur in
response to your
actions. 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 Exits Grid World Actions Deterministic Grid World Stochastic Grid World Markov Decision Processes ▪ An MDP is defined by: ▪ A set of states s S ▪ A set of actions a A ▪ A transition function T(s, a, s’) ▪ Probability that a from s leads to s’ ▪ i.e., P(s’| s, a) ▪ Also called the model or the dynamics ▪ A reward function R(s, a, s’)
▪ Sometimes just R(s) or R(s’) ▪ A start state ▪ Maybe a terminal state MDPs are non-deterministic search problems • One way to solve them is with expectimax search 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 ▪ This is just like search, where the successor function could only
depend on the current state (not the history) Andrey Markov
(1856-1922) Policies Optimal policy when R(s, a, s’) = -0.03
for all non-terminal s ▪ In deterministic single-agent search problems,
we wanted an optimal plan, or sequence of
actions, from start to a goal. ▪ For MDPs, we want an optimal policy *: S → A ▪ A policy gives an action for each state ▪ An optimal policy is one that maximizes
expected utility if followed ▪ An explicit policy defines a reflex agent ▪ Expectimax didn’t compute entire policies ▪ It computed the action for a single state only Which path would you take? Optimal Policies R(s) = -2.0R(s) = -0.4 R(s) = -0.03R(s) = -0.01 R(s) = the
“living reward” i.e., the
(negative)
reward that you
get when you
don’t exit. The agent
has a lot of
patience to
explore. Finding the
gem and
falling in to
the fire pit
gets the
same score The agent is
better of
dying ASAP! Example: Racing Example: Racing ▪ A robot car wants to travel far, quickly ▪ Three states: Cool, Warm, Overheated ▪ Two actions: Slow, Fast ▪ Going faster gets double reward Cool Warm Overheated Fast Fast Slow Slow 0.5
0.5
0.5
0.5
1.0
1.0
+1
+1
+1
+2
+2
-10 Racing Search Tree MDP Search Trees ▪ Each MDP state projects an expectimax-like search tree a s s’ s, a (s,a,s’) called a transition T(s,a,s’) = P(s’|s,a) R(s,a,s’) s,a,s’ s is a state (s, a) is a q- state Utilities of Sequences ▪ What preferences should an agent have over reward sequences? ▪ More or less? ▪ Now or later? Utilities of Sequences [1, 2, 2] [2, 3, 4]or [0, 0, 1] [1, 0, 0]or Discounting ▪ It’s reasonable to maximize the sum of rewards ▪ It’s also reasonable to prefer rewards now to rewards later ▪ One solution: values of rewards decay exponentially Worth Now Worth Next Step Worth In Two Steps Discounting ▪ How to discount? ▪ Each time we descend a level, we
multiply in the discount once ▪ Why discount? ▪ Sooner rewards probably do have
higher utility than later rewards ▪ Also helps our algorithms converge ▪ Example: discount of 0.5 ▪ U([1,2,3]) = 1*1 + 0.5*2 + 0.25*3 ▪ U([1,2,3]) < U([3,2,1]) Stationary Preferences ▪ Theorem: if we assume stationary preferences: ▪ Then: there are only two ways to define utilities ▪ Additive utility: ▪ Discounted utility: Quiz: Discounting ▪ Given: ▪ Quiz 1: For = 1, what is the optimal policy for
states b, c, and d? ▪ Quiz 2: For = 0.1, what is the optimal policy when
in state d? ▪ Quiz 3: For which are West and East equally good
when in state d? Actions: East, West, and Exit (only available in exit states a, e) Transitions: deterministic =
1 √10 Infinite Utilities?! ▪ Problem: What if the game lasts forever?
Do we get infinite rewards? ▪ Solutions: ▪ Finite horizon: (similar to depth-limited search) ▪ Terminate episodes after a fixed T steps (e.g. life) ▪ Gives nonstationary policies ( depends on time left) ▪ Discounting: use 0 < < 1 ▪ Smaller means smaller “horizon” – shorter term focus ▪ Absorbing state: guarantee that for every policy, a terminal state will eventually
be reached (like “overheated” for racing) Recap: Defining MDPs ▪ Markov decision processes: ▪ Set of states S ▪ Start state s0 ▪ Set of actions A ▪ Transitions P(s’|s,a) (or T(s,a,s’)) ▪ Rewards R(s,a,s’) (and discount ) ▪ MDP quantities so far: ▪ Policy = Choice of action for each state ▪ Utility = sum of (discounted) rewards a s s, a s,a,s’ s’ Solving MDPs Optimal Quantities ▪ The value (utility) of a state s: V*(s) = expected utility starting in s and
acting optimally ▪ The value (utility) of a q-state (s,a): Q*(s,a) = expected utility starting out
having taken action a from state s and
(thereafter) acting optimally ▪ The optimal policy: *(s) = optimal action from state s a s s’ s, a (s,a,s’) is a
transition s,a,s’ s is a
state (s, a) is a
q-state [Demo – gridworld values (L8D4)] Snapshot of Demo – Gridworld V Values Noise = 0.2 Discount = 0.9 Living reward = 0 Snapshot of Demo – Gridworld Q Values Noise = 0.2 Discount = 0.9 Living reward = 0 Values of States ▪ Fundamental operation: compute the (expectimax) value of a state ▪ Expected utility under optimal action ▪ Average sum of (discounted) rewards ▪ This is just what expectimax computed! ▪ Recursive definition of value: a s s, a s,a,s’ s’ We will revisit this: Bellman equations Racing Search Tree Racing Search Tree Racing Search Tree ▪ We’re doing way too much
work with expectimax! ▪ Problem: States are repeated
▪ Idea: Only compute needed
quantities once ▪ Problem: Tree goes on forever ▪ Idea: Do a depth-limited
computation, but with increasing
depths until change is small ▪ Note: deep parts of the tree
eventually don’t matter if γ < 1 Time-Limited Values ▪ Key idea: time-limited values ▪ Define Vk(s) to be the optimal value of s if the game ends
in k more time steps ▪ Equivalently, it’s what a depth-k expectimax would give from s [Demo – time-limited values (L8D6)] k=0 Noise = 0.2 Discount = 0.9 Living reward = 0 k=1 Noise = 0.2 Discount = 0.9 Living reward = 0 k=2 Noise = 0.2 Discount = 0.9 Living reward = 0 k=3 Noise = 0.2 Discount = 0.9 Living reward = 0 k=4 Noise = 0.2 Discount = 0.9 Living reward = 0 k=5 Noise = 0.2 Discount = 0.9 Living reward = 0 k=6 Noise = 0.2 Discount = 0.9 Living reward = 0 k=7 Noise = 0.2 Discount = 0.9 Living reward = 0 k=8 Noise = 0.2 Discount = 0.9 Living reward = 0 k=9 Noise = 0.2 Discount = 0.9 Living reward = 0 k=10 Noise = 0.2 Discount = 0.9 Living reward = 0 k=11 Noise = 0.2 Discount = 0.9 Living reward = 0 k=12 Noise = 0.2 Discount = 0.9 Living reward = 0 k=100 Noise = 0.2 Discount = 0.9 Living reward = 0 Computing Time-Limited Values Value Iteration Value Iteration ▪ Start with V0(s) = 0: no time steps left means an expected reward sum of zero ▪ Given vector of Vk(s) values, do one ply of expectimax from each state: ▪ Repeat until convergence ▪ Complexity of each iteration: O(S2A) ▪ Theorem: will converge to unique optimal values ▪ Basic idea: approximations get refined towards optimal values ▪ Policy may converge long before values do a Vk+1(s) s, a s,a,s’ Vk(s’) Example: Value Iteration 0
0
0 2
1
0 3.5
2.5
0 Assume no discount! Convergence* ▪ How do we know the Vk vectors are going to converge? ▪ Case 1: If the tree has maximum depth M, then VM holds
the actual untruncated values ▪ Case 2: If the discount is less than 1 ▪ Sketch: For any state Vk and Vk+1 can be viewed as depth
k+1 expectimax results in nearly identical search trees ▪ The difference is that on the bottom layer, Vk+1 has actual
rewards while Vk has zeros ▪ That last layer is at best all RMAX
▪ It is at worst RMIN
▪ But everything is discounted by γk that far out ▪ So Vk and Vk+1 are at most γk max|R| different ▪ So as k increases, the values converge Next Time:
MDP II: Policy-Based Methods and RL Intro 51作业君版权所有
