Q Learning
Presented by,
Shobitha AS
R22DE138
Blue and Pink Professional Business Strategy Presentation Page 1of 6-Q Learning is a reinforcement learning algorithm.
__ INTRODUCTION “T*2ins 9 tne the optimal actionsetection
policy for an agent in a Markov decision process
TO Q-LEARNING wo».
+ It learns by iteratively updating the Q-values of
state-action pairs based on the rewards received
from the environment.
Blue and Pink Professional Business Strategy Presentation Page 20f 6AN
ARNING
Q Learning is important in enabling
agents to learn and make decisions in
complex environments. It allows
agents to adapt to changing situations
and make decisions based on the
expected outcomes of their actions.
This is particularly useful in
applications such as robotics, where
agents need to navigate and interact
with their environment in a dynamic
and uncertain way
Blue and Pink Professional Business Strategy PresentationQ-Learning algorithm
Q(St, At) — Q(St Ae) + a[Revr + ymaraQ(Se11,4) — Q(S1, Ar)]
New Former Learning Immediate Discounted Estimate Former
Qvalue Quvalue Rate Reward optimal Q-value Qvalue
estimation estimation of next state estimation
TD Target
TD Error
Blue and Pink Professional Business Strategy Presentation Page 4 of 6Applications of OQ Learning
e e 7 e
GAME PLAYING — ROBOTICS angen
LEARNING CAN BE
Q LEARNING CAN BE PPLED TO GrrMize
USED INROBOTICS TO RESUS i
QUARNING HAS BEEN FRAN AUTONOMOUS fimecaeler ANB
SUCCESSFULLY ie RUENTSNONICATE (RES Donna
AE SroMatoers Wee ss SONAR
PLAYING, SUCH ASTIN AN TAS ochtay
THe PMiOUS Cage OF MurgowmenSMams Hutte Conrao.,
ALPHAGO, WHERE IT ENABLES THE ROBOT MANAGEMENT. iT
WAS USto TO TRAN ENABLES, THE ROEGT MANAGEMENT aald
THE Al AGENT TO PLAY AND. “ERROR ano Eres” octets
THE GaMe OF GO ATA Make “*"SECIsiONS 2uanns SMES
SUPERMUMAN LEVEL BASE on ncwanes —SRwHE-osrs
END BeNALTIES
Blue and Pink Professional Business Strategy Presentation Page 5 of 6Blue and Pink Professional Business Strategy Presentation Page 6 of 6