RL_online _presentation_1.ppt

Reinforcement Learning
Yijue Hou
What is learning?
 Learning takes place as a result of interaction
between an agent and the world, the idea
behind learning is that
 Percepts received by an agent should be used not
only for acting, but also for improving the agent’s
ability to behave optimally in the future to achieve
the goal.
Learning types
 Learning types
 Supervised learning:
a situation in which sample (input, output) pairs of the
function to be learned can be perceived or are given
 You can think it as if there is a kind teacher
 Reinforcement learning:
in the case of the agent acts on its environment, it
receives some evaluation of its action (reinforcement),
but is not told of which action is the correct one to
achieve its goal
Reinforcement learning
 Task
Learn how to behave successfully to achieve a
goal while interacting with an external
environment
 Learn via experiences!
 Examples
 Game playing: player knows whether it win or lose,
but not know how to move at each step
 Control: a traffic system can measure the delay of
cars, but not know how to decrease it.
RL is learning from interaction
RL model
 Each percept(e) is enough to determine the
State(the state is accessible)
 The agent can decompose the Reward component
from a percept.
 The agent task: to find a optimal policy, mapping
states to actions, that maximize long-run measure
of the reinforcement
 Think of reinforcement as reward
 Can be modeled as MDP model!
Review of MDP model
 MDP model <S,T,A,R>
Agent
Environment
State
Reward
Action
s0
r0
a0
s1
a1
r1
s2
a2
r2
s3
• S– set of states
• A– set of actions
• T(s,a,s’) = P(s’|s,a)– the
probability of transition from
s to s’ given action a
• R(s,a)– the expected reward
for taking action a in state s




'
'
)
'
,
,
(
)
'
,
,
(
)
,
(
)
'
,
,
(
)
,
|
'
(
)
,
(
s
s
s
a
s
r
s
a
s
T
a
s
R
s
a
s
r
a
s
s
P
a
s
R
Model based v.s.Model free
approaches
 But, we don’t know anything about the environment
model—the transition function T(s,a,s’)
 Here comes two approaches
 Model based approach RL:
learn the model, and use it to derive the optimal policy.
e.g Adaptive dynamic learning(ADP) approach
 Model free approach RL:
derive the optimal policy without learning the model.
e.g LMS and Temporal difference approach
 Which one is better?
Passive learning v.s. Active
learning
 Passive learning
 The agent imply watches the world going by and
tries to learn the utilities of being in various states
 Active learning
 The agent not simply watches, but also acts
Example environment
Passive learning scenario
 The agent see the the sequences of state
transitions and associate rewards
 The environment generates state transitions and the
agent perceive them
e.g (1,1) (1,2) (1,3) (2,3) (3,3) (4,3)[+1]
(1,1)(1,2) (1,3) (1,2) (1,3) (1,2) (1,1) (2,1)
(3,1) (4,1) (4,2)[-1]
 Key idea: updating the utility value using the
given training sequences.
Passive leaning scenario
LMS updating
 Reward to go of a state
the sum of the rewards from that state until a
terminal state is reached
 Key: use observed reward to go of the state as
the direct evidence of the actual expected utility
of that state
 Learning utility function directly from sequence
example
LMS updating
function LMS-UPDATE (U, e, percepts, M, N ) return an updated U
if TERMINAL?[e] then
{ reward-to-go  0
for each ei in percepts (starting from end) do
s = STATE[ei]
reward-to-go  reward-to-go + REWARS[ei]
U[s] = RUNNING-AVERAGE (U[s], reward-to-go, N[s])
end
}
function RUNNING-AVERAGE (U[s], reward-to-go, N[s] )
U[s] = [ U[s] * (N[s] – 1) + reward-to-go ] / N[s]
LMS updating algorithm in
passive learning
 Drawback:
 The actual utility of a state is constrained to be probability- weighted
average of its successor’s utilities.
 Converge very slowly to correct utilities values (requires a lot of sequences)
 for our example, >1000!
Temporal difference method in
passive learning
 TD(0) key idea:
 adjust the estimated utility value of the current state based on its
immediately reward and the estimated value of the next state.
 The updating rule
 is the learning rate parameter
 Only when is a function that decreases as the number of times
a state has been visited increased, then can U(s)converge to the
correct value.
))
(
)
'
(
)
(
(
)
(
)
( s
U
s
U
s
R
s
U
s
U 


 


The TD learning curve
(4,3)
(2,3)
(2,2)
(1,1)
(3,1)
(4,1)
(4,2)
Adaptive dynamic programming(ADP)
in passive learning
 Different with LMS and TD method(model free
approaches)
 ADP is a model based approach!
 The updating rule for passive learning
 However, in an unknown environment, T is not given,
the agent must learn T itself by experiences with the
environment.
 How to learn T?
))
'
(
)
'
,
(
(
)
'
,
(
)
(
'
s
U
s
s
r
s
s
T
s
U
s


 
ADP learning curves
(4,3)
(3,3)
(2,3)
(1,1)
(3,1)
(4,1)
(4,2)
Active learning
 An active agent must consider
 what actions to take?
 what their outcomes maybe(both on learning and receiving the
rewards in the long run)?
 Update utility equation
 Rule to chose action
))
'
(
)
'
,
,
(
)
,
(
(
max
arg
'
s
U
s
a
s
T
a
s
R
a
s
a


 
))
'
(
)
'
,
,
(
)
,
(
(
max
)
(
'
s
U
s
a
s
T
a
s
R
s
U
s
a


 
Active ADP algorithm
For each s, initialize U(s) , T(s,a,s’) and R(s,a)
Initialize s to current state that is perceived
Loop forever
{
Select an action a and execute it (using current model R and T) using
Receive immediate reward r and observe the new state s’
Using the transition tuple <s,a,s’,r> to update model R and T (see further)
For all the sate s, update U(s) using the updating rule
s = s’
}
))
'
(
)
'
,
,
(
)
,
(
(
max
arg
'
s
U
s
a
s
T
a
s
R
a
s
a


 
))
'
(
)
'
,
,
(
)
,
(
(
max
)
(
'
s
U
s
a
s
T
a
s
R
s
U
s
a


 
How to learn model?
 Use the transition tuple <s, a, s’, r> to learn T(s,a,s’) and
R(s,a). That’s supervised learning!
 Since the agent can get every transition (s, a, s’,r) directly, so take
(s,a)/s’ as an input/output example of the transition probability
function T.
 Different techniques in the supervised learning(see further reading
for detail)
 Use r and T(s,a,s’) to learn R(s,a)


'
)
'
,
,
(
)
,
(
s
r
s
a
s
T
a
s
R
ADP approach pros and cons
 Pros:
 ADP algorithm converges far faster than LMS and Temporal
learning. That is because it use the information from the the model
of the environment.
 Cons:
 Intractable for large state space
 In each step, update U for all states
 Improve this by prioritized-sweeping (see further reading for detail)
Another model free method–
TD-Q learning
 Define Q-value function
 Q-value function updating rule
<*>
 Key idea of TD-Q learning
 Combined with temporal difference approach
 The updating rule
 Rule to chose the action to take
)
,
(
max
)
( a
s
Q
s
U
a

))
'
(
)
'
,
,
(
)
,
(
(
max
)
(
'
s
U
s
a
s
T
a
s
R
s
U
s
a


 
)
'
(
)
'
,
,
(
)
,
(
)
,
(
'
s
U
s
a
s
T
a
s
R
a
s
Q
s


 
)
'
,
'
(
max
)
'
,
,
(
)
,
(
)
,
(
'
'
a
s
Q
s
a
s
T
a
s
R
a
s
Q
s
a


 
))
,
(
)
'
,
'
(
max
(
)
,
(
)
,
(
'
a
s
Q
a
s
Q
r
a
s
Q
a
s
Q
a



 

)
,
(
max
arg a
s
Q
a
a

TD-Q learning agent algorithm
For each pair (s, a), initialize Q(s,a)
Observe the current state s
Loop forever
{
Select an action a and execute it
Receive immediate reward r and observe the new state s’
Update Q(s,a)
s=s’
}
)
,
(
max
arg a
s
Q
a
a

))
,
(
)
'
,
'
(
max
(
)
,
(
)
,
(
'
a
s
Q
a
s
Q
r
a
s
Q
a
s
Q
a



 

 An action has two kinds of outcome
 Gain rewards on the current experience
tuple (s,a,s’)
 Affect the percepts received, and hence
the ability of the agent to learn
Exploration problem in Active
learning
Exploration problem in Active
learning
 A trade off when choosing action between
 its immediately good(reflected in its current utility estimates using the
what we have learned)
 its long term good(exploring more about the environment help it to
behave optimally in the long run)
 Two extreme approaches
 “wacky”approach: acts randomly, in the hope that it will eventually
explore the entire environment.
 “greedy”approach: acts to maximize its utility using current model
estimate
See Figure 20.10
 Just like human in the real world! People need to decide between
 Continuing in a comfortable existence
 Or striking out into the unknown in the hopes of discovering a new
and better life
Exploration problem in Active
learning
 One kind of solution: the agent should be more wacky when it has
little idea of the environment, and more greedy when it has a
model that is close to being correct
 In a given state, the agent should give some weight to actions that it
has not tried very often.
 While tend to avoid actions that are believed to be of low utility
 Implemented by exploration function f(u,n):
 assigning a higher utility estimate to relatively unexplored action state
pairs
 Chang the updating rule of value function to
 U+ denote the optimistic estimate of the utility
))
,
(
),
'
(
)
'
,
,
(
(
)
,
(
(
max
)
(
'
s
a
N
s
U
s
a
s
T
f
a
s
r
s
U
s
a




 
Exploration problem in Active
learning
 One kind of definition of f(u,n)
if n< Ne
u otherwise
 is an optimistic estimate of the best possible reward
obtainable in any state
 The agent will try each action-state pair(s,a) at least Ne times
 The agent will behave initially as if there were wonderful rewards
scattered all over around– optimistic .

)
,
( n
u
f

R


R
Generalization in
Reinforcement Learning
 So far we assumed that all the functions learned by
the agent are (U, T, R,Q) are tabular forms—
i.e.. It is possible to enumerate state and action
spaces.
 Use generalization techniques to deal with large state
or action space.
 Function approximation techniques
Genetic algorithm and Evolutionary
programming
 Start with a set of individuals
 Apply selection and reproduction operators to “evolve” an individual that is
successful (measured by a fitness function)
Genetic algorithm and Evolutionary
programming
 Imagine the individuals as agent functions
 Fitness function as performance measure or reward
function
 No attempt made to learn the relationship the
rewards and actions taken by an agent
 Simply searches directly in the individual space to
find one that maximizes the fitness functions
Genetic algorithm and Evolutionary
programming
 Represent an individual as a binary string(each bit of the string is called a gene)
 Selection works like this: if individual X scores twice as high as Y on the fitness
function, then X is twice likely to be selected for reproduction than Y is
 Reproduction is accomplished by cross-over and mutation
Thank you!
1 de 34

Recomendados

reiniforcement learning.ppt por
reiniforcement learning.pptreiniforcement learning.ppt
reiniforcement learning.pptcharusharma165
11 visualizações34 slides
Reinforcement Learning.ppt por
Reinforcement Learning.pptReinforcement Learning.ppt
Reinforcement Learning.pptPOOJASHREEC1
22 visualizações34 slides
Reinforcement learning 7313 por
Reinforcement learning 7313Reinforcement learning 7313
Reinforcement learning 7313Slideshare
13.7K visualizações30 slides
RL.ppt por
RL.pptRL.ppt
RL.pptAzharJamil15
30 visualizações16 slides
An efficient use of temporal difference technique in Computer Game Learning por
An efficient use of temporal difference technique in Computer Game LearningAn efficient use of temporal difference technique in Computer Game Learning
An efficient use of temporal difference technique in Computer Game LearningPrabhu Kumar
70 visualizações18 slides
REINFORCEMENT LEARNING por
REINFORCEMENT LEARNINGREINFORCEMENT LEARNING
REINFORCEMENT LEARNINGpradiprahul
235 visualizações54 slides

Mais conteúdo relacionado

Similar a RL_online _presentation_1.ppt

Lecture notes por
Lecture notesLecture notes
Lecture notesbutest
302 visualizações21 slides
reinforcement-learning-141009013546-conversion-gate02.pdf por
reinforcement-learning-141009013546-conversion-gate02.pdfreinforcement-learning-141009013546-conversion-gate02.pdf
reinforcement-learning-141009013546-conversion-gate02.pdfVaishnavGhadge1
117 visualizações64 slides
Reinforcement Learning por
Reinforcement LearningReinforcement Learning
Reinforcement LearningSVijaylakshmi
207 visualizações81 slides
lecture_21.pptx - PowerPoint Presentation por
lecture_21.pptx - PowerPoint Presentationlecture_21.pptx - PowerPoint Presentation
lecture_21.pptx - PowerPoint Presentationbutest
398 visualizações19 slides
Reinforcement learning por
Reinforcement learning Reinforcement learning
Reinforcement learning Chandra Meena
44.7K visualizações64 slides
Intro rl por
Intro rlIntro rl
Intro rlRonald Teo
417 visualizações46 slides

Similar a RL_online _presentation_1.ppt(20)

Lecture notes por butest
Lecture notesLecture notes
Lecture notes
butest302 visualizações
reinforcement-learning-141009013546-conversion-gate02.pdf por VaishnavGhadge1
reinforcement-learning-141009013546-conversion-gate02.pdfreinforcement-learning-141009013546-conversion-gate02.pdf
reinforcement-learning-141009013546-conversion-gate02.pdf
VaishnavGhadge1117 visualizações
Reinforcement Learning por SVijaylakshmi
Reinforcement LearningReinforcement Learning
Reinforcement Learning
SVijaylakshmi207 visualizações
lecture_21.pptx - PowerPoint Presentation por butest
lecture_21.pptx - PowerPoint Presentationlecture_21.pptx - PowerPoint Presentation
lecture_21.pptx - PowerPoint Presentation
butest398 visualizações
Reinforcement learning por Chandra Meena
Reinforcement learning Reinforcement learning
Reinforcement learning
Chandra Meena44.7K visualizações
Intro rl por Ronald Teo
Intro rlIntro rl
Intro rl
Ronald Teo417 visualizações
Reinfrocement Learning por Natan Katz
Reinfrocement LearningReinfrocement Learning
Reinfrocement Learning
Natan Katz175 visualizações
RL_Dr.SNR Final ppt for Presentation 28.05.2021.pptx por deeplearning6
RL_Dr.SNR Final ppt for Presentation 28.05.2021.pptxRL_Dr.SNR Final ppt for Presentation 28.05.2021.pptx
RL_Dr.SNR Final ppt for Presentation 28.05.2021.pptx
deeplearning610 visualizações
AUTOMATIC TRANSFER RATE ADJUSTMENT FOR TRANSFER REINFORCEMENT LEARNING por gerogepatton
AUTOMATIC TRANSFER RATE ADJUSTMENT FOR TRANSFER REINFORCEMENT LEARNINGAUTOMATIC TRANSFER RATE ADJUSTMENT FOR TRANSFER REINFORCEMENT LEARNING
AUTOMATIC TRANSFER RATE ADJUSTMENT FOR TRANSFER REINFORCEMENT LEARNING
gerogepatton20 visualizações
CS3013 -MACHINE LEARNING.pptx por logesswarisrinivasan
CS3013 -MACHINE LEARNING.pptxCS3013 -MACHINE LEARNING.pptx
CS3013 -MACHINE LEARNING.pptx
logesswarisrinivasan18 visualizações
Reinforcement learning 1.pdf por ssuserd27779
Reinforcement learning 1.pdfReinforcement learning 1.pdf
Reinforcement learning 1.pdf
ssuserd2777994 visualizações
Cs221 rl por darwinrlo
Cs221 rlCs221 rl
Cs221 rl
darwinrlo307 visualizações
Reinforcement Learning Guide For Beginners por gokulprasath06
Reinforcement Learning Guide For BeginnersReinforcement Learning Guide For Beginners
Reinforcement Learning Guide For Beginners
gokulprasath06296 visualizações
Lecture 1 - introduction.pdf por NamanJain758248
Lecture 1 - introduction.pdfLecture 1 - introduction.pdf
Lecture 1 - introduction.pdf
NamanJain7582484 visualizações
Matineh Shaker, Artificial Intelligence Scientist, Bonsai at MLconf SF 2017 por MLconf
Matineh Shaker, Artificial Intelligence Scientist, Bonsai at MLconf SF 2017Matineh Shaker, Artificial Intelligence Scientist, Bonsai at MLconf SF 2017
Matineh Shaker, Artificial Intelligence Scientist, Bonsai at MLconf SF 2017
MLconf924 visualizações
M Harmon RL Tutorial por Mance Harmon
M Harmon RL TutorialM Harmon RL Tutorial
M Harmon RL Tutorial
Mance Harmon272 visualizações
Deep Reinforcement Learning Through Policy Optimization, John Schulman, OpenAI por Jack Clark
Deep Reinforcement Learning Through Policy Optimization, John Schulman, OpenAIDeep Reinforcement Learning Through Policy Optimization, John Schulman, OpenAI
Deep Reinforcement Learning Through Policy Optimization, John Schulman, OpenAI
Jack Clark2.9K visualizações
Hierarchical Reinforcement Learning por ahmad bassiouny
Hierarchical Reinforcement LearningHierarchical Reinforcement Learning
Hierarchical Reinforcement Learning
ahmad bassiouny904 visualizações
Q_Learning.ppt por AyushGiri27
Q_Learning.pptQ_Learning.ppt
Q_Learning.ppt
AyushGiri2713 visualizações

Último

CASE IH JX1070V Tractor Service Repair Manual Instant Download.pdf por
CASE IH JX1070V Tractor Service Repair Manual Instant Download.pdfCASE IH JX1070V Tractor Service Repair Manual Instant Download.pdf
CASE IH JX1070V Tractor Service Repair Manual Instant Download.pdfruan49549560
8 visualizações32 slides
JOHN DEERE 5210 TRACTOR Service Repair Manual Instant Download (TM1716).pdf por
JOHN DEERE 5210 TRACTOR Service Repair Manual Instant Download (TM1716).pdfJOHN DEERE 5210 TRACTOR Service Repair Manual Instant Download (TM1716).pdf
JOHN DEERE 5210 TRACTOR Service Repair Manual Instant Download (TM1716).pdfqianxu35543635
24 visualizações36 slides
Volvo EC300D NL EC300DNL Excavator Service Repair Manual Instant Download.pdf por
Volvo EC300D NL EC300DNL Excavator Service Repair Manual Instant Download.pdfVolvo EC300D NL EC300DNL Excavator Service Repair Manual Instant Download.pdf
Volvo EC300D NL EC300DNL Excavator Service Repair Manual Instant Download.pdfdai20nao
5 visualizações20 slides
Caterpillar Cat 931B TRACK LOADER (Prefix 29Y) Service Repair Manual Instant ... por
Caterpillar Cat 931B TRACK LOADER (Prefix 29Y) Service Repair Manual Instant ...Caterpillar Cat 931B TRACK LOADER (Prefix 29Y) Service Repair Manual Instant ...
Caterpillar Cat 931B TRACK LOADER (Prefix 29Y) Service Repair Manual Instant ...jiuliu14811471
14 visualizações25 slides
Audience profile (1).pptx por
Audience profile (1).pptxAudience profile (1).pptx
Audience profile (1).pptxMollyBrown86
7 visualizações3 slides
One of the Most Reliable Refrigerated Transport in Brisbane and Gold Coast por
One of the Most Reliable Refrigerated Transport in Brisbane and Gold CoastOne of the Most Reliable Refrigerated Transport in Brisbane and Gold Coast
One of the Most Reliable Refrigerated Transport in Brisbane and Gold CoastJD Refrigerated Transport
7 visualizações6 slides

Último(20)

CASE IH JX1070V Tractor Service Repair Manual Instant Download.pdf por ruan49549560
CASE IH JX1070V Tractor Service Repair Manual Instant Download.pdfCASE IH JX1070V Tractor Service Repair Manual Instant Download.pdf
CASE IH JX1070V Tractor Service Repair Manual Instant Download.pdf
ruan495495608 visualizações
JOHN DEERE 5210 TRACTOR Service Repair Manual Instant Download (TM1716).pdf por qianxu35543635
JOHN DEERE 5210 TRACTOR Service Repair Manual Instant Download (TM1716).pdfJOHN DEERE 5210 TRACTOR Service Repair Manual Instant Download (TM1716).pdf
JOHN DEERE 5210 TRACTOR Service Repair Manual Instant Download (TM1716).pdf
qianxu3554363524 visualizações
Volvo EC300D NL EC300DNL Excavator Service Repair Manual Instant Download.pdf por dai20nao
Volvo EC300D NL EC300DNL Excavator Service Repair Manual Instant Download.pdfVolvo EC300D NL EC300DNL Excavator Service Repair Manual Instant Download.pdf
Volvo EC300D NL EC300DNL Excavator Service Repair Manual Instant Download.pdf
dai20nao5 visualizações
Caterpillar Cat 931B TRACK LOADER (Prefix 29Y) Service Repair Manual Instant ... por jiuliu14811471
Caterpillar Cat 931B TRACK LOADER (Prefix 29Y) Service Repair Manual Instant ...Caterpillar Cat 931B TRACK LOADER (Prefix 29Y) Service Repair Manual Instant ...
Caterpillar Cat 931B TRACK LOADER (Prefix 29Y) Service Repair Manual Instant ...
jiuliu1481147114 visualizações
Audience profile (1).pptx por MollyBrown86
Audience profile (1).pptxAudience profile (1).pptx
Audience profile (1).pptx
MollyBrown867 visualizações
One of the Most Reliable Refrigerated Transport in Brisbane and Gold Coast por JD Refrigerated Transport
One of the Most Reliable Refrigerated Transport in Brisbane and Gold CoastOne of the Most Reliable Refrigerated Transport in Brisbane and Gold Coast
One of the Most Reliable Refrigerated Transport in Brisbane and Gold Coast
JD Refrigerated Transport7 visualizações
Volvo EC360C L EC360CL Excavator Service Repair Manual Instant Download.pdf por lunrizan628
Volvo EC360C L EC360CL Excavator Service Repair Manual Instant Download.pdfVolvo EC360C L EC360CL Excavator Service Repair Manual Instant Download.pdf
Volvo EC360C L EC360CL Excavator Service Repair Manual Instant Download.pdf
lunrizan62814 visualizações
294196410-investigatoy-project-on-application-of-biotechnology.docx por maiyadeengupta94
294196410-investigatoy-project-on-application-of-biotechnology.docx294196410-investigatoy-project-on-application-of-biotechnology.docx
294196410-investigatoy-project-on-application-of-biotechnology.docx
maiyadeengupta949 visualizações
Volvo EC140C L (EC140CL) Excavator Service Repair Manual Instant Download.pdf por dai20nao
Volvo EC140C L (EC140CL) Excavator Service Repair Manual Instant Download.pdfVolvo EC140C L (EC140CL) Excavator Service Repair Manual Instant Download.pdf
Volvo EC140C L (EC140CL) Excavator Service Repair Manual Instant Download.pdf
dai20nao5 visualizações
JOHN DEERE 316 LAWN GARDEN TRACTOR Service Repair Manual Instant Download (TM... por bu760639265959
JOHN DEERE 316 LAWN GARDEN TRACTOR Service Repair Manual Instant Download (TM...JOHN DEERE 316 LAWN GARDEN TRACTOR Service Repair Manual Instant Download (TM...
JOHN DEERE 316 LAWN GARDEN TRACTOR Service Repair Manual Instant Download (TM...
bu76063926595915 visualizações
Caterpillar Cat 325F L Excavator (Prefix RBW) Service Repair Manual Instant D... por qianxu35543635
Caterpillar Cat 325F L Excavator (Prefix RBW) Service Repair Manual Instant D...Caterpillar Cat 325F L Excavator (Prefix RBW) Service Repair Manual Instant D...
Caterpillar Cat 325F L Excavator (Prefix RBW) Service Repair Manual Instant D...
qianxu3554363511 visualizações
JCB 536-60 Tier 4 Telescopic Handler Service Repair Manual SN from 2902000 on... por caijianwi2
JCB 536-60 Tier 4 Telescopic Handler Service Repair Manual SN from 2902000 on...JCB 536-60 Tier 4 Telescopic Handler Service Repair Manual SN from 2902000 on...
JCB 536-60 Tier 4 Telescopic Handler Service Repair Manual SN from 2902000 on...
caijianwi25 visualizações
Volvo L30B Compact Wheel Loader Service Repair Manual Instant Download.pdf por lunrizan628
Volvo L30B Compact Wheel Loader Service Repair Manual Instant Download.pdfVolvo L30B Compact Wheel Loader Service Repair Manual Instant Download.pdf
Volvo L30B Compact Wheel Loader Service Repair Manual Instant Download.pdf
lunrizan6288 visualizações
Volvo L90F Wheel Loader Service Repair Manual Instant Download.pdf por lunrizan628
Volvo L90F Wheel Loader Service Repair Manual Instant Download.pdfVolvo L90F Wheel Loader Service Repair Manual Instant Download.pdf
Volvo L90F Wheel Loader Service Repair Manual Instant Download.pdf
lunrizan6286 visualizações
Volvo L90E Wheel Loader Service Repair Manual Instant Download.pdf por fapanhe306271
Volvo L90E Wheel Loader Service Repair Manual Instant Download.pdfVolvo L90E Wheel Loader Service Repair Manual Instant Download.pdf
Volvo L90E Wheel Loader Service Repair Manual Instant Download.pdf
fapanhe3062719 visualizações
Volvo EC25 Compact Excavator Service Repair Manual Instant Download.pdf por lunrizan628
Volvo EC25 Compact Excavator Service Repair Manual Instant Download.pdfVolvo EC25 Compact Excavator Service Repair Manual Instant Download.pdf
Volvo EC25 Compact Excavator Service Repair Manual Instant Download.pdf
lunrizan6286 visualizações
John Deere 4710 Compact Utility Tractor Service Repair Manual Instant Downloa... por qianxu35543635
John Deere 4710 Compact Utility Tractor Service Repair Manual Instant Downloa...John Deere 4710 Compact Utility Tractor Service Repair Manual Instant Downloa...
John Deere 4710 Compact Utility Tractor Service Repair Manual Instant Downloa...
qianxu3554363524 visualizações
Caterpillar Cat 930H Wheel Loader (Prefix DHC) Service Repair Manual Instant ... por jiuliu14811471
Caterpillar Cat 930H Wheel Loader (Prefix DHC) Service Repair Manual Instant ...Caterpillar Cat 930H Wheel Loader (Prefix DHC) Service Repair Manual Instant ...
Caterpillar Cat 930H Wheel Loader (Prefix DHC) Service Repair Manual Instant ...
jiuliu1481147116 visualizações
JCB JZ140 Tracked Excavator Parts Catalogue Manual Instant Download (SN 01137... por niekanggu15
JCB JZ140 Tracked Excavator Parts Catalogue Manual Instant Download (SN 01137...JCB JZ140 Tracked Excavator Parts Catalogue Manual Instant Download (SN 01137...
JCB JZ140 Tracked Excavator Parts Catalogue Manual Instant Download (SN 01137...
niekanggu155 visualizações
Caterpillar Cat 930R WHEEL LOADER (Prefix 57Z) Service Repair Manual Instant ... por jiuliu14811471
Caterpillar Cat 930R WHEEL LOADER (Prefix 57Z) Service Repair Manual Instant ...Caterpillar Cat 930R WHEEL LOADER (Prefix 57Z) Service Repair Manual Instant ...
Caterpillar Cat 930R WHEEL LOADER (Prefix 57Z) Service Repair Manual Instant ...
jiuliu1481147117 visualizações

RL_online _presentation_1.ppt

  • 2. What is learning?  Learning takes place as a result of interaction between an agent and the world, the idea behind learning is that  Percepts received by an agent should be used not only for acting, but also for improving the agent’s ability to behave optimally in the future to achieve the goal.
  • 3. Learning types  Learning types  Supervised learning: a situation in which sample (input, output) pairs of the function to be learned can be perceived or are given  You can think it as if there is a kind teacher  Reinforcement learning: in the case of the agent acts on its environment, it receives some evaluation of its action (reinforcement), but is not told of which action is the correct one to achieve its goal
  • 4. Reinforcement learning  Task Learn how to behave successfully to achieve a goal while interacting with an external environment  Learn via experiences!  Examples  Game playing: player knows whether it win or lose, but not know how to move at each step  Control: a traffic system can measure the delay of cars, but not know how to decrease it.
  • 5. RL is learning from interaction
  • 6. RL model  Each percept(e) is enough to determine the State(the state is accessible)  The agent can decompose the Reward component from a percept.  The agent task: to find a optimal policy, mapping states to actions, that maximize long-run measure of the reinforcement  Think of reinforcement as reward  Can be modeled as MDP model!
  • 7. Review of MDP model  MDP model <S,T,A,R> Agent Environment State Reward Action s0 r0 a0 s1 a1 r1 s2 a2 r2 s3 • S– set of states • A– set of actions • T(s,a,s’) = P(s’|s,a)– the probability of transition from s to s’ given action a • R(s,a)– the expected reward for taking action a in state s     ' ' ) ' , , ( ) ' , , ( ) , ( ) ' , , ( ) , | ' ( ) , ( s s s a s r s a s T a s R s a s r a s s P a s R
  • 8. Model based v.s.Model free approaches  But, we don’t know anything about the environment model—the transition function T(s,a,s’)  Here comes two approaches  Model based approach RL: learn the model, and use it to derive the optimal policy. e.g Adaptive dynamic learning(ADP) approach  Model free approach RL: derive the optimal policy without learning the model. e.g LMS and Temporal difference approach  Which one is better?
  • 9. Passive learning v.s. Active learning  Passive learning  The agent imply watches the world going by and tries to learn the utilities of being in various states  Active learning  The agent not simply watches, but also acts
  • 11. Passive learning scenario  The agent see the the sequences of state transitions and associate rewards  The environment generates state transitions and the agent perceive them e.g (1,1) (1,2) (1,3) (2,3) (3,3) (4,3)[+1] (1,1)(1,2) (1,3) (1,2) (1,3) (1,2) (1,1) (2,1) (3,1) (4,1) (4,2)[-1]  Key idea: updating the utility value using the given training sequences.
  • 13. LMS updating  Reward to go of a state the sum of the rewards from that state until a terminal state is reached  Key: use observed reward to go of the state as the direct evidence of the actual expected utility of that state  Learning utility function directly from sequence example
  • 14. LMS updating function LMS-UPDATE (U, e, percepts, M, N ) return an updated U if TERMINAL?[e] then { reward-to-go  0 for each ei in percepts (starting from end) do s = STATE[ei] reward-to-go  reward-to-go + REWARS[ei] U[s] = RUNNING-AVERAGE (U[s], reward-to-go, N[s]) end } function RUNNING-AVERAGE (U[s], reward-to-go, N[s] ) U[s] = [ U[s] * (N[s] – 1) + reward-to-go ] / N[s]
  • 15. LMS updating algorithm in passive learning  Drawback:  The actual utility of a state is constrained to be probability- weighted average of its successor’s utilities.  Converge very slowly to correct utilities values (requires a lot of sequences)  for our example, >1000!
  • 16. Temporal difference method in passive learning  TD(0) key idea:  adjust the estimated utility value of the current state based on its immediately reward and the estimated value of the next state.  The updating rule  is the learning rate parameter  Only when is a function that decreases as the number of times a state has been visited increased, then can U(s)converge to the correct value. )) ( ) ' ( ) ( ( ) ( ) ( s U s U s R s U s U       
  • 17. The TD learning curve (4,3) (2,3) (2,2) (1,1) (3,1) (4,1) (4,2)
  • 18. Adaptive dynamic programming(ADP) in passive learning  Different with LMS and TD method(model free approaches)  ADP is a model based approach!  The updating rule for passive learning  However, in an unknown environment, T is not given, the agent must learn T itself by experiences with the environment.  How to learn T? )) ' ( ) ' , ( ( ) ' , ( ) ( ' s U s s r s s T s U s    
  • 20. Active learning  An active agent must consider  what actions to take?  what their outcomes maybe(both on learning and receiving the rewards in the long run)?  Update utility equation  Rule to chose action )) ' ( ) ' , , ( ) , ( ( max arg ' s U s a s T a s R a s a     )) ' ( ) ' , , ( ) , ( ( max ) ( ' s U s a s T a s R s U s a    
  • 21. Active ADP algorithm For each s, initialize U(s) , T(s,a,s’) and R(s,a) Initialize s to current state that is perceived Loop forever { Select an action a and execute it (using current model R and T) using Receive immediate reward r and observe the new state s’ Using the transition tuple <s,a,s’,r> to update model R and T (see further) For all the sate s, update U(s) using the updating rule s = s’ } )) ' ( ) ' , , ( ) , ( ( max arg ' s U s a s T a s R a s a     )) ' ( ) ' , , ( ) , ( ( max ) ( ' s U s a s T a s R s U s a    
  • 22. How to learn model?  Use the transition tuple <s, a, s’, r> to learn T(s,a,s’) and R(s,a). That’s supervised learning!  Since the agent can get every transition (s, a, s’,r) directly, so take (s,a)/s’ as an input/output example of the transition probability function T.  Different techniques in the supervised learning(see further reading for detail)  Use r and T(s,a,s’) to learn R(s,a)   ' ) ' , , ( ) , ( s r s a s T a s R
  • 23. ADP approach pros and cons  Pros:  ADP algorithm converges far faster than LMS and Temporal learning. That is because it use the information from the the model of the environment.  Cons:  Intractable for large state space  In each step, update U for all states  Improve this by prioritized-sweeping (see further reading for detail)
  • 24. Another model free method– TD-Q learning  Define Q-value function  Q-value function updating rule <*>  Key idea of TD-Q learning  Combined with temporal difference approach  The updating rule  Rule to chose the action to take ) , ( max ) ( a s Q s U a  )) ' ( ) ' , , ( ) , ( ( max ) ( ' s U s a s T a s R s U s a     ) ' ( ) ' , , ( ) , ( ) , ( ' s U s a s T a s R a s Q s     ) ' , ' ( max ) ' , , ( ) , ( ) , ( ' ' a s Q s a s T a s R a s Q s a     )) , ( ) ' , ' ( max ( ) , ( ) , ( ' a s Q a s Q r a s Q a s Q a       ) , ( max arg a s Q a a 
  • 25. TD-Q learning agent algorithm For each pair (s, a), initialize Q(s,a) Observe the current state s Loop forever { Select an action a and execute it Receive immediate reward r and observe the new state s’ Update Q(s,a) s=s’ } ) , ( max arg a s Q a a  )) , ( ) ' , ' ( max ( ) , ( ) , ( ' a s Q a s Q r a s Q a s Q a      
  • 26.  An action has two kinds of outcome  Gain rewards on the current experience tuple (s,a,s’)  Affect the percepts received, and hence the ability of the agent to learn Exploration problem in Active learning
  • 27. Exploration problem in Active learning  A trade off when choosing action between  its immediately good(reflected in its current utility estimates using the what we have learned)  its long term good(exploring more about the environment help it to behave optimally in the long run)  Two extreme approaches  “wacky”approach: acts randomly, in the hope that it will eventually explore the entire environment.  “greedy”approach: acts to maximize its utility using current model estimate See Figure 20.10  Just like human in the real world! People need to decide between  Continuing in a comfortable existence  Or striking out into the unknown in the hopes of discovering a new and better life
  • 28. Exploration problem in Active learning  One kind of solution: the agent should be more wacky when it has little idea of the environment, and more greedy when it has a model that is close to being correct  In a given state, the agent should give some weight to actions that it has not tried very often.  While tend to avoid actions that are believed to be of low utility  Implemented by exploration function f(u,n):  assigning a higher utility estimate to relatively unexplored action state pairs  Chang the updating rule of value function to  U+ denote the optimistic estimate of the utility )) , ( ), ' ( ) ' , , ( ( ) , ( ( max ) ( ' s a N s U s a s T f a s r s U s a      
  • 29. Exploration problem in Active learning  One kind of definition of f(u,n) if n< Ne u otherwise  is an optimistic estimate of the best possible reward obtainable in any state  The agent will try each action-state pair(s,a) at least Ne times  The agent will behave initially as if there were wonderful rewards scattered all over around– optimistic .  ) , ( n u f  R   R
  • 30. Generalization in Reinforcement Learning  So far we assumed that all the functions learned by the agent are (U, T, R,Q) are tabular forms— i.e.. It is possible to enumerate state and action spaces.  Use generalization techniques to deal with large state or action space.  Function approximation techniques
  • 31. Genetic algorithm and Evolutionary programming  Start with a set of individuals  Apply selection and reproduction operators to “evolve” an individual that is successful (measured by a fitness function)
  • 32. Genetic algorithm and Evolutionary programming  Imagine the individuals as agent functions  Fitness function as performance measure or reward function  No attempt made to learn the relationship the rewards and actions taken by an agent  Simply searches directly in the individual space to find one that maximizes the fitness functions
  • 33. Genetic algorithm and Evolutionary programming  Represent an individual as a binary string(each bit of the string is called a gene)  Selection works like this: if individual X scores twice as high as Y on the fitness function, then X is twice likely to be selected for reproduction than Y is  Reproduction is accomplished by cross-over and mutation