Deep Q-Learning/Deep Q-Network (DQN) Explained | Python Pytorch Deep Reinforcement Learning

hey everyone welcome to my deep Q

learning tutorial here's what to expect

from this video since deep Q learning is

a little bit complicated to explain I'm

going to use the Frozen Lake reinforcement

learning environment it's a very simple

environment so I'm going to do a quick

intro on how the environment works or

also quickly answer the question of why

we need reinforcement learning on such a

simple environment next to navigate the

environment we need to use the Epsilon

greedy algorithm both Q learning and

deep Q learning uses the same

algorithm after we know how to navigate

the environment we're going to take a

look at the differences of the output

between Q learning and deep Q learning

in Q learning we're training a q table

in deep Q learning we're training a deep

Q Network we'll work through how the Q

table is trained then we can see how it

is different from training a deep Q

Network in training a deep Q Network we

also need a technique called EXP

experience replay and after we have an

idea of how deep Q learning works I'm

going to walk through the code and also

run it and demo

it just in case you're not familiar with

how this environment Works quick recap

so the goal is to get the learning agent

to figure out how to get to the goal on

the bottom right the actions that the

agent can take is left internity we're

going to represent that as zero down is

one right is two up is three in this

case if the agent tries to go left or up

it's just going to stay in place so in

general if it tries to go off the grid

it's going to stay in the same spot

internally we're going to represent each

state as this one is zero this one is 1

2

3 uh 4 5 6 all the way to

14 15 so 16 States total each attempt at

navigating the map is considered one

episode the episode ends if the the

agent falls into one of these holes or

it reaches the goal when it reaches the

goal it will receive a reward of one all

other states have no reward or penalty

okay so that's how this environment

works at this point you might be

wondering why we would need

reinforcement learning to solve such a

simple map why not just use a

pathfinding algorithm there is a Twist

to this environment there is a flag

called is

Slippery When set to

true the agent doesn't always execute

the action that it intends to for

example if the agent wants to go right

there's only a oneir chance that it will

execute this action and there's a two3

chance of it going in an adjacent

Direction now with slippery turn on

pathf finding algorithm will not be able

to solve this in fact I'm not even sure

what algorithm can that's where

reinforcement learning comes in if I

don't know how to solve this I can just

give the agent some incentive and let it

figure out how to solve this so that's

where reinforcement learning

shines in terms of how the agent

navigates the map both deep Q learning

and Q learning uses the Epsilon greedy

algorithm basically we start with a

variable call Epsilon equal and set it

equal to one and then we generate a

random number if the random number is

less than Epsilon we pick a random

action otherwise we pick the best action

that we know of at the moment and at the

end of each episode we'll decrease

Epsilon by a little bit at a time so

essentially we start off with 100%

random exploration and then eventually

near the end of training we're going to

be always selecting the best action to

take before we jump into the details of

how the training Works between QQ

learning versus deep Q learning let's

take a look at what the outcome looks

like for Q learning the output the

output of the training is a q table

which is nothing more than a

two-dimensional array consisting of 16

states by four actions after training

the whole table is going to be filled

with Q values for example it might look

something like

this so in this case the agent is at

state zero we look at the table and see

what the maximum value is in this case

it's to go right so this is the

prescribed action over at Deep Q

learning the output is a deep Q Network

which is actually nothing more than a

regular feedforward Nur Network this is

actually what it looks like but I think

we should switch back to the simplified

view the input layer is going to have 16

nodes the output layer is going to have

four

nodes the way that we Center input into

the input layer is like this if the

agent is at state zero we're going to

set the first note to one and everything

else

zero if the agent is at State one then

the first Noe is zero and the second Noe

is one and everything else is zero so

this is called one hot encoding just do

one more if it's at the

last if we want to put in state 15 then

everything all zeros and state 15 is one

so that's how the input works with the

input the output that gets calculated

are Q values now the Q values are not

going to look like what's in the Q table

but it might be something similar let me

throw some numbers in here

okay so same thing for this particular

input the best action is the highest Q

value when training a neural network

essentially we're trying to train the

weights associated with each one of

these lines in the neural network and

the bias for all the hidden layers now

how do we know how many hidden layers we

need in this case one layer was enough

but you can certainly add more if

necessary and how many nodes do we need

in the hidden layer I try 16 and it's

able to solve the map so I'm sticking

with that but you can certainly increase

or decrease this number to see what

happens okay so that's the differences

between the output of Q learning versus

deep Q

learning as the agent navigates the map

it is using the Q learning formula to

calculate the Q values and update the Q

table now the formula might look a

little bit scary but it's actually not

that bad let's work through some

examples let's say our agent is in state

14 and it's going right to get to State

15 we're calculating Q of State 14

Action 2 which is this cell the current

value of that cell is zero because we

initialize the whole Q table to zero at

the beginning plus the learning rate the

learning rate is a hyperparameter that

we can set as an example I'll just put

in

0.01

times the reward we get a reward of one

because we reach the goal plus the

discount Factor another parameter that

we set I'll just use 0.9 times the max Q

value of the new state so the max Q

value of State 15

since State 15 is a terminal state it

will never get anything other than zeros

in this

table Max is zero essentially these two

are gone subtracted by the same thing

that we had

here okay so work through the math this

is just

0.01 so we get 0.01

here now let's do another one really

quickly agent is here take a right Q of

13 Going to the right 13 is also all

zeros this is the one we're

updating here is zero plus the learning

rate times a reward there is no reward

plus the discount Factor times Max of

the new state Max of State

14 Max of State 14 is

0.01 subtracted by again it's

zero this is equal to

0.9 okay it's actually pretty

straightforward now how the heck does

this formula help find the PATH so if we

train enough the theory is that this

number is going to be really close to

one and then the states next to

it it's going to be 0.9

something and then the states adjacent

to that it's probably going to be 0.8

[Music]

something and if we keep

going we can see that a path is

is actually two

paths are possible so mathematically

this is how the path is

found over at Deep Q learning the

formula is going to look like this we

set Q equal to the reward if the new

state is a terminal State otherwise we

set it

to this

part

of the KE learning

formula let's see how the formula is

going be used in

training for DEQ learning we actually

need two neural networks let me walk

through the steps the network on the

left is called the policy Network this

is the network that we're going to do

the training on the one on the right is

called the target Network the target

network is the one that makes use of the

DQ learning formula now let's walk

through the steps of training step one

is going to be creation of this policy

Network step two we make a copy of the

policy Network into the target Network

so basically we're copying this the

weights and the bias over here so both

networks are identical step number three

the agent navigates the map as usual

let's say the agent is here in state 14

and it's going into State 15 step three

is just navigation now step four we

input State 14 remember how we have to

encode the input it's going to look like

this 0 1 2

3 12

13 State 14 15 the input is going to

look like this as you know neural

networks when it's created it comes with

a random set of weights and bias so with

this input we're actually going to get

an output even though these values are

pretty much

meaningless so we might get some stuff

that looks

like I'm just putting in some random

numbers okay so these Q values are

meaningless and as a reminder this is

the left action down right and up okay

step five

we do the same thing we take the exact

same input State 14 and send it into the

target Network which will also

calculate the exact same numbers

because the target network is the same

as the policy Network

currently step six this is where we

calculate the Q value for State 14 we're

taking the action of

two is equal to since we're going into

State 15 it is a terminal

State we set it equal

to

1 Step seven step seven is to set the

target input State 14 output action two

is this

node we take the value that we

calculated up on step six and we replace

place the Q value in the output step

number eight we take the target Q

values we use it to train the policy

Network so this value is the one that's

really going to change as you know with

neural networks it doesn't go straight

to one it's going to go toward that

direction so maybe maybe it'll go to

0.01 but if you repeat the training many

many times it will approach One Step

nine is to repeat the whole thing

again of course we're not going to

create the policy Network we not going

to make a copy of it so what we're

repeating is steps three through 8 step

10 after a certain number of steps or

episodes we're going to sync the policy

network with the target Network which

means we're going to make them identical

by copying the weight and biases from

the policy over to the Target Network

after syncing the networks we're going

to repeat nine again and then repeat 10

until training is done okay so that's

generally how deep que Learning Works um

that might have been a little bit

complicated so maybe rewind the video

watch it again and make sure you

understand what's

happening to effectively train a neural

network we need to randomize the

training data that we send into the

neural network however if you remember

from steps three and four the agent

takes an action and then we're sending

that training data into the neur network

so the question is how do we how do we

randomize the order of a single sample

but that's where experience replay comes

in we need a step 3A where we memorize

the agent's experience as the agent

navigates the map rest storing the state

that it was in what action it took the

new state that it reached if there was a

reward or not and and if the new state

is a terminal state or not we take that

and insert it into the memory and the

memory is nothing more than a python

deck how a deck works is that as the

deck gets full whatever at the end gets

purged we need a step 3B this is the

replay step this is where we take say 30

random samples from the deck and then

pass it on to step four for training so

that's what experience replay

is before we jump into the code if you

have problems installing gymnasium

especially on Windows I got a video for

that I'm not going to walk through the Q

learning code but if you are interested

in that jump to my Q learning Code

walkthrough video after watching this

one also if you have zero experience

with neural networks you can check out

this basic tutorial on neuron networks

you o need pytorch so head over to PYT

to.org and get that installed the first

thing that we do is create the class to

represent our deep Q Network I'm calling

it dqn as mentioned before a deep Q

network is nothing more than a feed

forward node Network so there's really

nothing special about it here I'm using

pretty standard way of creating a node

network using pytorch if you look up any

pytorch tutorial you'll probably see

something just just like this so since

this is not a py torch tutorial I'm not

going to spend too much time explaining

how py torch works for this class we

have to inherit the neuron network

module which requires us to implement

two functions the inet function and the

forward function in the init function

I'm passing in the number of nodes in my

input State hidden layer and output

State back at our diagram we're going to

have 16 input States as mentioned before

I'm using 16 in the hidden layer that's

something that you can adjust yourself

and in the output layer four

notes we declare the hidden layer which

has 16 going into 16 and then the output

layer 16 going into four and then the

forward function X is the training data

set and we're sending the training data

through the neuron Network again this is

pretty common pytorch

code next we need a class to represent

the replay

memory so it's this portion that we're

implementing right

now in the init function we'll pass in a

max length and then create the python

deck in the append function we're going

to append the transition the transition

is this tupo here State action new state

reward and

terminated these sample function will

return a random sample of whatever size

we want from the memory and then the

link function simply Returns the length

of the

memory the fen L dql class is where

we're going to do our

training we set the learning rate and

the discount Factor uh those are part of

the Q learning formula and these are

values that you can adjust and play

around

with the network sync rate it's the

number of steps the agent takes before

syncing the policy and Target

Network that's the setting for step 10

Where We sync the policy and the target

Network we set the replay memory size to

1,000 and the replay memory sample size

to 32 these are also numbers that you

can uh play around with next is the loss

function and the optimizer these two are

py torch variables for the loss function

I'm simply using the mean square error

function for the optimizer will

initialize that at a later time this

actions list is to Simply map the action

numbers into letters for

printing in the train function we can

specify how many episodes we want to

train the agent whether we want to

render the map on screen and what do we

want to turn on the slippery

flag we'll instantiate the frosen lake

environment and then create a variable

to store the number of states this is

going to be 16 inst store the number of

actions this is going to be four here we

initialize Epsilon to one and we

instantiate the replay memory this is

going to be size of uh 1,000 now we

create the policy DEQ

Network this is step one create the

policy

Network we also create a Target Network

and copy the weights and bias from the

policy Network into the target Network

so that

is Step number two making the Target and

policy Network

identical before training we'll do a

print out of the policy Network just so

we can compare it to the end result next

we initialize the optimizer that we

declared

earlier and we're simply using the atom

Optimizer passing in the learning rate

we'll use this rewards per episode list

to keep track of the rewards collected

per

episode we'll also use this Epsilon

history list to keep track of Epsilon

decaying over

time we'll use step count to keep track

of the number of steps taken this is

used to determine when to sync the

policy and Target Network again so that

is for step 10 the

syncing next we will Loop through the

number of episodes that were specified

when calling the train function we'll

initialize the map so the agent starts

at state zero we'll initialize the

terminated and truncated flag terminated

is when when the agent falls into the

hole or reaches the goal truncated is

when the agent takes more than 200

actions on such a small map 4x4 uh this

is probably never going to occur but

we'll keep this anyway now we have a

while loop checking for the terminated

and truncated flag keep looping until

these conditions are met this part is

basically the Epsilon greedy algorithm

if a random number is less than Epsilon

we'll just pick a random action

otherwise we'll use the policy Network

to calculate the set of Q values take

the maximum of that extract that item

and that's going to be the best action

so we have a function here called state

to dqn input let's jump down there

remember that we have to take the state

and encode it that's what this function

does this type of encoding

here okay let's go back

with P torch if we want to perform a

prediction we should call this torch. no

grads so that it doesn't calculate the

stuff needed for training after

selecting the either a random action or

the best action we'll call the step

function to execute the action when we

take that action it's going to return

the new state whether there is a reward

or not whether it's a terminal state or

we got truncated we take all that

information

and put it into our memory so that is

Step 3A when we did the Epsilon greedy

that was the step

three after executing the action we're

resetting the state equal to the new

state we increment our step counter if

we received an award put it on our list

now we'll check the memory to see if we

have enough training data to to do

optimization on also we want to check if

we have collected at least one reward if

we haven't collected any rewards there's

really no point in optimizing the

network if those conditions are met we

use memory. sample and we pass in uh the

batch size which was 32 and we get a

batch of training data out of the memory

we'll pass that training data into the

optimized function along with the policy

Network and the target Network

let's jump over to the optimize

function this first line is just looking

at the policy Network and getting the

number of input noes we expect this to

be 16 the current qist and Target

qist so this is the current

cus this is the target

cus what I'm about to describe now is

Step 3B rep playing the experience

so to replay the experience We're

looping through the training data inside

the mini batch let's jump down here for

a moment so this is Step number four

we're taking the states and passing it

into the policy Network to calculate the

current list of Q

values step number four the output is

the list of Q

values step number five we pass in the

same thing to the Target

Network step five we get the Q values

out here which should be the same as the

Q values from the policy

Network step number six is up here so

reminder of what it looks like step six

is using this formula

here so that's what we have here if

terminated if terminated just return the

reward otherwise use that that second

formula so now that we have a Target

we'll go to step seven step seven is

taking the output of Step six and

replacing the respective Q

value and that's what we're doing here

we're replacing the Q value of that

particular action with the Target that

was calculated up

above step number eight step eight is to

take the target values and use that to

train the current Q values

so that's what we have here we're using

the loss function pass in the current

set of Q values plus the target set of Q

values and then this is just standard Pi

torch code to optimize the policy

Network now step nine step nine is to

repeat steps 3 to 8 starting from

navigation to optimizing the

network

basically that's just continuing this

inner while loop of stepping through the

states and this outer for Loop of

stepping through each

episode step 10 is where we sync the

policy Network and the target

Network and we do that down here if the

number of steps taken is greater than

the network sync rate that we set then

we copy the policy Network into the

target Network and then we reset the

step counter also after each episode we

should be decaying the Epsilon value

after syncing the

network we repeat Step n again which is

repeating basically the navigation and

training

again so it's just going to go back up

here and do this all over again all the

way until we have finished the number of

episodes so that was training after

training we close the environment we can

save the policy or the weights and bias

into a file I'm hardcoding it here you

can definitely make this more Dynamic

here I'm creating a new graph and I'm

basically graphing the rewards collected

per episode also I'm graphing the

Epsilon history here after graphing I'm

saving the graphs into an image all

right so that was the train function we

have the talked about the optimize

function let me fold

that we already looked at that now the

test function is going to run the fren

Lake environment with the policy that we

learned from the train function we can

also pass in the number of episodes and

whether we want to turn on slippery or

not so the rest of the code is going to

look pretty similar to what we had in

the train function we instantiate the

environment get the number of states and

action s 16 and 4 here we declar the

policy Network load from the file that

we saved from training this is just P

torch code to switch the policy Network

to prediction mode or evaluation mode

rather than training mode we'll print

the train policy and then we'll Loop

over the episodes we set the agent up

top you've seen this before keep looping

until the agent gets terminated or

truncated here we're selecting the best

action out of the policy Network and

executing the action and then close the

environment so ideally when we run this

we're going to see the agent navigate

the map and reach the goal however if

slippery is on there is no guarantee

that the agent is going to solve it in

one try it might take a couple of tries

for the agent to solve the

map all right down at the main function

we create an instance of the fen Lake

dql class first first we're going to try

non-slippery we'll train it for a th

times or 1 th000 episodes we'll run the

test four times just because the map

pops up and goes away when the agent

gets to the goal so um we want to run it

a couple times so we actually can see it

on the screen okay I'm just hitting

contrl

F5 all right training is done and looks

like the agent is able to solve the map

we also have a print out of the policy

that it's learned I print it in a way

that matches up with the grid let's jump

back to this in a second let's go to our

files so after training we have a new

graph on the left side that's the number

of rewards collected over the 1,000

episodes as you can see over time it's

improving and finally at the end it's

getting a lot of rewards on the right

side we can see Epsilon decaying

starting from one slowly slowly down to

zero also another file that gets created

is it's a binary file so we can't really

display it but it contains the weights

and biases of the policy

Network so if we want to do the test

again we don't need the training we can

comment out the training line and just

run this

again and we can see the agent solve the

map

we can look at what policy the agent

learned the way this is printed it

matches up with the map so at state zero

the best action was to go right which is

State one at State one the best action

is going to go right again at state two

go down at St six go down again at St 10

go down again at State 14

go to the right

okay so it was right right down down

down right so remember earlier that

there were two possible path the one

that we actually learned plus the one

going down because Epsilon greedy has a

Randomness Factor whether the agent

learns the top path or the bottom path

is just somewhat random if weet train

again maybe it'll go down the next time

now let's turn the slippery Factor

on uncomment the training

line now with slippery on we expect the

agent to fail a lot more often or to

fall in the holes a lot more often let

me just triple the number of um

training and let me do 10 times for

testing okay I'm running this again

again with slippery Turn on there's no

guarantee that after training the agent

is going to be able to pass every

episode let's make it come

up you see it trying to go to the bottom

right but because of slippery it just

failed right there but there you go it

was able to solve it let's look at the

graph the results compared to the

non-slippery surface is significantly

worse also you want to be careful

getting stuck in spots like these which

means if I was unlucky enough to set my

training episodes to maybe 2200 and it

gets stuck here which means it learned a

bad policy it won't be able to solve the

map so in your training if you're

finding that the agent is not able to

solve the map when slippery is turned on

it might be because of things like these

okay that concludes our deep Q learning

tutorial I love to hear feedback from

you was the explanation easy to

understand what can I improve on and

what other topics are you interested in