Welcome to the fourth project of the Machine Learning Engineer Nanodegree! In this notebook, template code has already been provided for you to aid in your analysis of the Smartcab and your implemented learning algorithm. You will not need to modify the included code beyond what is requested. There will be questions that you must answer which relate to the project and the visualizations provided in the notebook. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide in agent.py.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. In addition, Markdown cells can be edited by typically double-clicking the cell to enter edit mode.
In this project, you will work towards constructing an optimized Q-Learning driving agent that will navigate a Smartcab through its environment towards a goal. Since the Smartcab is expected to drive passengers from one location to another, the driving agent will be evaluated on two very important metrics: Safety and Reliability. A driving agent that gets the Smartcab to its destination while running red lights or narrowly avoiding accidents would be considered unsafe. Similarly, a driving agent that frequently fails to reach the destination in time would be considered unreliable. Maximizing the driving agent's safety and reliability would ensure that Smartcabs have a permanent place in the transportation industry.
Safety and Reliability are measured using a letter-grade system as follows:
| Grade | Safety | Reliability |
|---|---|---|
| A+ | Agent commits no traffic violations, and always chooses the correct action. |
Agent reaches the destination in time for 100% of trips. |
| A | Agent commits few minor traffic violations, such as failing to move on a green light. |
Agent reaches the destination on time for at least 90% of trips. |
| B | Agent commits frequent minor traffic violations, such as failing to move on a green light. |
Agent reaches the destination on time for at least 80% of trips. |
| C | Agent commits at least one major traffic violation, such as driving through a red light. |
Agent reaches the destination on time for at least 70% of trips. |
| D | Agent causes at least one minor accident, such as turning left on green with oncoming traffic. |
Agent reaches the destination on time for at least 60% of trips. |
| F | Agent causes at least one major accident, such as driving through a red light with cross-traffic. |
Agent fails to reach the destination on time for at least 60% of trips. |
To assist evaluating these important metrics, you will need to load visualization code that will be used later on in the project. Run the code cell below to import this code which is required for your analysis.
In [2]:
# Import the visualization code
import visuals as vs
# Pretty display for notebooks
%matplotlib inline
Before starting to work on implementing your driving agent, it's necessary to first understand the world (environment) which the Smartcab and driving agent work in. One of the major components to building a self-learning agent is understanding the characteristics about the agent, which includes how the agent operates. To begin, simply run the agent.py agent code exactly how it is -- no need to make any additions whatsoever. Let the resulting simulation run for some time to see the various working components. Note that in the visual simulation (if enabled), the white vehicle is the Smartcab.
In a few sentences, describe what you observe during the simulation when running the default agent.py agent code. Some things you could consider:
Hint: From the /smartcab/ top-level directory (where this notebook is located), run the command
'python smartcab/agent.py'
Answer: The Smartcab doesn't move at all during the simulation. No action is initiated in the code. Since the only action performed is 'None', the agent receives a positive reward when the light is red (as it complies with traffic rule), and gets a negative reward when the light is green and there is no conflicting oncoming traffic (since the smartcan does not move).
In addition to understanding the world, it is also necessary to understand the code itself that governs how the world, simulation, and so on operate. Attempting to create a driving agent would be difficult without having at least explored the "hidden" devices that make everything work. In the /smartcab/ top-level directory, there are two folders: /logs/ (which will be used later) and /smartcab/. Open the /smartcab/ folder and explore each Python file included, then answer the following question.
agent.py Python file, choose three flags that can be set and explain how they change the simulation.environment.py Python file, what Environment class function is called when an agent performs an action?simulator.py Python file, what is the difference between the 'render_text()' function and the 'render()' function?planner.py Python file, will the 'next_waypoint() function consider the North-South or East-West direction first?Answer:
agent.py file Flag 1: learning in agent creation. If set to true, the agent will use Q-learning to improve its decision making, and its behavior will not be random.
Flag 2: Alpha: Learning rate which determines to what extent the newly acquired information will override the old information. A factor of 0 will make the agent not learn anything, while a factor of 1 would make the agent consider only the most recent information
Flag 3: enforce_deadline: if set to a given value, the duration of the trip will be relevant and the agent's behavior will be affected by the time to destination (if the agent has not reached the target within the deadline, the agent will be penalized). If enforce_deadline is set to false, the duration of the trip does not matter.
environment.py file The class function 'act' is called when an agent performs an action.
simulator.py file The render_text() is a function for a non-GUI render display of the simulation whereas render() is the GUI display of the simulation.
planner.py file The 'next_waypoint() function will consider the East-West direction first.
The first step to creating an optimized Q-Learning driving agent is getting the agent to actually take valid actions. In this case, a valid action is one of None, (do nothing) 'Left' (turn left), 'Right' (turn right), or 'Forward' (go forward). For your first implementation, navigate to the 'choose_action()' agent function and make the driving agent randomly choose one of these actions. Note that you have access to several class variables that will help you write this functionality, such as 'self.learning' and 'self.valid_actions'. Once implemented, run the agent file and simulation briefly to confirm that your driving agent is taking a random action each time step.
To obtain results from the initial simulation, you will need to adjust following flags:
'enforce_deadline' - Set this to True to force the driving agent to capture whether it reaches the destination in time.'update_delay' - Set this to a small value (such as 0.01) to reduce the time between steps in each trial.'log_metrics' - Set this to True to log the simluation results as a .csv file in /logs/.'n_test' - Set this to '10' to perform 10 testing trials.Optionally, you may disable to the visual simulation (which can make the trials go faster) by setting the 'display' flag to False. Flags that have been set here should be returned to their default setting when debugging. It is important that you understand what each flag does and how it affects the simulation!
Once you have successfully completed the initial simulation (there should have been 20 training trials and 10 testing trials), run the code cell below to visualize the results. Note that log files are overwritten when identical simulations are run, so be careful with what log file is being loaded! Run the agent.py file after setting the flags from projects/smartcab folder instead of projects/smartcab/smartcab.
In [2]:
# Load the 'sim_no-learning' log file from the initial simulation results
vs.plot_trials('sim_no-learning.csv')
Using the visualization above that was produced from your initial simulation, provide an analysis and make several observations about the driving agent. Be sure that you are making at least one observation about each panel present in the visualization. Some things you could consider:
Answer:
Relatively large ratio of bad decisions (between 35% and 40%). Around a fourth of these are leading to accidents.
The rate of reliability remains low at around 10%. This makes sense given that the agent is driving randomly.
The agent receives positive rewards when performing a correct action such as not moving when a light is red and is penalized for bad actions. The graph displaying rolling average rewards shows negative values around -5, consistent with the fact that a large ratio of actions that the agent takes are bad actions for which it gets penalized.
As the number of trials increases, the outcome of results does not change significantly. The ratio of bad actions is relatively stable, so is the average reward.
Given the large ratio of bad actions and the reliability rate close to zero, this Smartcab would be consider neither safe, nor reliable for its passengers.
The second step to creating an optimized Q-learning driving agent is defining a set of states that the agent can occupy in the environment. Depending on the input, sensory data, and additional variables available to the driving agent, a set of states can be defined for the agent so that it can eventually learn what action it should take when occupying a state. The condition of 'if state then action' for each state is called a policy, and is ultimately what the driving agent is expected to learn. Without defining states, the driving agent would never understand which action is most optimal -- or even what environmental variables and conditions it cares about!
Inspecting the 'build_state()' agent function shows that the driving agent is given the following data from the environment:
'waypoint', which is the direction the Smartcab should drive leading to the destination, relative to the Smartcab's heading.'inputs', which is the sensor data from the Smartcab. It includes 'light', the color of the light.'left', the intended direction of travel for a vehicle to the Smartcab's left. Returns None if no vehicle is present.'right', the intended direction of travel for a vehicle to the Smartcab's right. Returns None if no vehicle is present.'oncoming', the intended direction of travel for a vehicle across the intersection from the Smartcab. Returns None if no vehicle is present.'deadline', which is the number of actions remaining for the Smartcab to reach the destination before running out of time.Answer: 'Waypoint' is relevant for efficiency, since the agent should reach the destination within the deadline and should therefore seek to consider actions that are leading him into this direction. Deadline however is not relevant as its actions won't be influenced by it (and subsequently the optimality of the action does not depend on the deadline).
Light, left, right and oncoming are relevant to safety, since they are signals from the environment enabling the agent to assess which actions are permitted and which are not. For instance, if the light is red, the agent should either not move or move right if there is no oncoming traffic. If the light is green and the waypoint indicates left but there is oncoming traffic from cars across the intersection intending to cross it in a straight line, the agent should not perform the action of turning left, which would lead to an accident.
When defining a set of states that the agent can occupy, it is necessary to consider the size of the state space. That is to say, if you expect the driving agent to learn a policy for each state, you would need to have an optimal action for every state the agent can occupy. If the number of all possible states is very large, it might be the case that the driving agent never learns what to do in some states, which can lead to uninformed decisions. For example, consider a case where the following features are used to define the state of the Smartcab:
('is_raining', 'is_foggy', 'is_red_light', 'turn_left', 'no_traffic', 'previous_turn_left', 'time_of_day').
How frequently would the agent occupy a state like (False, True, True, True, False, False, '3AM')? Without a near-infinite amount of time for training, it's doubtful the agent would ever learn the proper action!
If a state is defined using the features you've selected from Question 4, what would be the size of the state space? Given what you know about the evironment and how it is simulated, do you think the driving agent could learn a policy for each possible state within a reasonable number of training trials?
Hint: Consider the combinations of features to calculate the total number of states!
Answer:
For each feature, here are all possible states:
Waypoint=[left, right, forward] (3 states)
Light=[Red, Green]
Left=[left, right, forward, None]
Right=[left, right, forward, None]
Oncoming=[left, right, forward, None]
This gives us a state space of size 384 (3x4x4x4x2). This is large space that would most likely require a very large number of training trials. In the remainder, I only consider the following features: Waypoint, Light, Oncoming.
For your second implementation, navigate to the 'build_state()' agent function. With the justification you've provided in Question 4, you will now set the 'state' variable to a tuple of all the features necessary for Q-Learning. Confirm your driving agent is updating its state by running the agent file and simulation briefly and note whether the state is displaying. If the visual simulation is used, confirm that the updated state corresponds with what is seen in the simulation.
Note: Remember to reset simulation flags to their default setting when making this observation!
The third step to creating an optimized Q-Learning agent is to begin implementing the functionality of Q-Learning itself. The concept of Q-Learning is fairly straightforward: For every state the agent visits, create an entry in the Q-table for all state-action pairs available. Then, when the agent encounters a state and performs an action, update the Q-value associated with that state-action pair based on the reward received and the interative update rule implemented. Of course, additional benefits come from Q-Learning, such that we can have the agent choose the best action for each state based on the Q-values of each state-action pair possible. For this project, you will be implementing a decaying, $\epsilon$-greedy Q-learning algorithm with no discount factor. Follow the implementation instructions under each TODO in the agent functions.
Note that the agent attribute self.Q is a dictionary: This is how the Q-table will be formed. Each state will be a key of the self.Q dictionary, and each value will then be another dictionary that holds the action and Q-value. Here is an example:
{ 'state-1': {
'action-1' : Qvalue-1,
'action-2' : Qvalue-2,
...
},
'state-2': {
'action-1' : Qvalue-1,
...
},
...
}Furthermore, note that you are expected to use a decaying $\epsilon$ (exploration) factor. Hence, as the number of trials increases, $\epsilon$ should decrease towards 0. This is because the agent is expected to learn from its behavior and begin acting on its learned behavior. Additionally, The agent will be tested on what it has learned after $\epsilon$ has passed a certain threshold (the default threshold is 0.01). For the initial Q-Learning implementation, you will be implementing a linear decaying function for $\epsilon$.
To obtain results from the initial Q-Learning implementation, you will need to adjust the following flags and setup:
'enforce_deadline' - Set this to True to force the driving agent to capture whether it reaches the destination in time.'update_delay' - Set this to a small value (such as 0.01) to reduce the time between steps in each trial.'log_metrics' - Set this to True to log the simluation results as a .csv file and the Q-table as a .txt file in /logs/.'n_test' - Set this to '10' to perform 10 testing trials.'learning' - Set this to 'True' to tell the driving agent to use your Q-Learning implementation.In addition, use the following decay function for $\epsilon$:
$$ \epsilon_{t+1} = \epsilon_{t} - 0.05, \hspace{10px}\textrm{for trial number } t$$If you have difficulty getting your implementation to work, try setting the 'verbose' flag to True to help debug. Flags that have been set here should be returned to their default setting when debugging. It is important that you understand what each flag does and how it affects the simulation!
Once you have successfully completed the initial Q-Learning simulation, run the code cell below to visualize the results. Note that log files are overwritten when identical simulations are run, so be careful with what log file is being loaded!
In [6]:
# Load the 'sim_default-learning' file from the default Q-Learning simulation
vs.plot_trials('sim_default-learning.csv')
Using the visualization above that was produced from your default Q-Learning simulation, provide an analysis and make observations about the driving agent like in Question 3. Note that the simulation should have also produced the Q-table in a text file which can help you make observations about the agent's learning. Some additional things you could consider:
Answer: The driving agent made 20 training trials before testing. This is consistent with an initial epsilon of 1 and the decay function of -0.05 at each trial. The parameters panel of the visualization is accurately representing the decaying function (decreasing linearly from 1 to 0 over the 20 trials). As the number of training trials increase the number of bad actions decreases (total percentage decreases from around 43% to around 14%). In parallel, the reward increased from -7 to -1 (remaining negative though). When applying learning, both the reliability and safety increase compared to the basic driving agent. Results vary substantially from one simulation to the next however, reflecting the high weight put on the exploration and the short duration of the training.
The third step to creating an optimized Q-Learning agent is to perform the optimization! Now that the Q-Learning algorithm is implemented and the driving agent is successfully learning, it's necessary to tune settings and adjust learning paramaters so the driving agent learns both safety and efficiency. Typically this step will require a lot of trial and error, as some settings will invariably make the learning worse. One thing to keep in mind is the act of learning itself and the time that this takes: In theory, we could allow the agent to learn for an incredibly long amount of time; however, another goal of Q-Learning is to transition from experimenting with unlearned behavior to acting on learned behavior. For example, always allowing the agent to perform a random action during training (if $\epsilon = 1$ and never decays) will certainly make it learn, but never let it act. When improving on your Q-Learning implementation, consider the impliciations it creates and whether it is logistically sensible to make a particular adjustment.
To obtain results from the initial Q-Learning implementation, you will need to adjust the following flags and setup:
'enforce_deadline' - Set this to True to force the driving agent to capture whether it reaches the destination in time.'update_delay' - Set this to a small value (such as 0.01) to reduce the time between steps in each trial.'log_metrics' - Set this to True to log the simluation results as a .csv file and the Q-table as a .txt file in /logs/.'learning' - Set this to 'True' to tell the driving agent to use your Q-Learning implementation.'optimized' - Set this to 'True' to tell the driving agent you are performing an optimized version of the Q-Learning implementation.Additional flags that can be adjusted as part of optimizing the Q-Learning agent:
'n_test' - Set this to some positive number (previously 10) to perform that many testing trials.'alpha' - Set this to a real number between 0 - 1 to adjust the learning rate of the Q-Learning algorithm.'epsilon' - Set this to a real number between 0 - 1 to adjust the starting exploration factor of the Q-Learning algorithm.'tolerance' - set this to some small value larger than 0 (default was 0.05) to set the epsilon threshold for testing.Furthermore, use a decaying function of your choice for $\epsilon$ (the exploration factor). Note that whichever function you use, it must decay to 'tolerance' at a reasonable rate. The Q-Learning agent will not begin testing until this occurs. Some example decaying functions (for $t$, the number of trials):
You may also use a decaying function for $\alpha$ (the learning rate) if you so choose, however this is typically less common. If you do so, be sure that it adheres to the inequality $0 \leq \alpha \leq 1$.
If you have difficulty getting your implementation to work, try setting the 'verbose' flag to True to help debug. Flags that have been set here should be returned to their default setting when debugging. It is important that you understand what each flag does and how it affects the simulation!
Once you have successfully completed the improved Q-Learning simulation, run the code cell below to visualize the results. Note that log files are overwritten when identical simulations are run, so be careful with what log file is being loaded!
NOTE TO REVIEWER: The below corresponds to the updated improved learning (with input['right'] and input['left'] included).
In [37]:
# Load the 'sim_improved-learning' file from the improved Q-Learning simulation
vs.plot_trials('sim_improved-learning.csv')
Using the visualization above that was produced from your improved Q-Learning simulation, provide a final analysis and make observations about the improved driving agent like in Question 6. Questions you should answer:
Answer: I tried several decaying functions for epsilon, one of which was the following: epsilon(t)=exp(-0.01 * t). Additionally, in order to reduce the influence of small upticks in bad actions towards the end of the trial (which could lead to overshooting by the agent), I also apply a decaying function for the learning rate (alpha=epsilon/3). I set the tolerance at 0.008.
It tool around 600 trials before the agent begins testing. The resulting Q-Learner is much more performant than the default Q-learner, with very good safety and reliability ratings.
I also implemented another linear decaying (with a decaying factor of -0.006), which leads to very good results (A/A+ ratings for both safety and reliability). In that case, it took around 350 trials before the agent began testing.
Sometimes, the answer to the important question "what am I trying to get my agent to learn?" only has a theoretical answer and cannot be concretely described. Here, however, you can concretely define what it is the agent is trying to learn, and that is the U.S. right-of-way traffic laws. Since these laws are known information, you can further define, for each state the Smartcab is occupying, the optimal action for the driving agent based on these laws. In that case, we call the set of optimal state-action pairs an optimal policy. Hence, unlike some theoretical answers, it is clear whether the agent is acting "incorrectly" not only by the reward (penalty) it receives, but also by pure observation. If the agent drives through a red light, we both see it receive a negative reward but also know that it is not the correct behavior. This can be used to your advantage for verifying whether the policy your driving agent has learned is the correct one, or if it is a suboptimal policy.
Provide a few examples (using the states you've defined) of what an optimal policy for this problem would look like. Afterwards, investigate the 'sim_improved-learning.txt' text file to see the results of your improved Q-Learning algorithm. For each state that has been recorded from the simulation, is the policy (the action with the highest value) correct for the given state? Are there any states where the policy is different than what would be expected from an optimal policy? Provide an example of a state and all state-action rewards recorded, and explain why it is the correct policy.
Answer:
Here are a few example of optimal policy: If waypoint is forward and the light is green, agent should go forward. If the waypoint is right and the light is red, the agent should turn right unless a car is coming from the left. If waypoint is left and the light is green, the agent should turn left unless there is oncoming traffic going forward.
Examples of state/optimal policies could be:
We can look into a couple of examples from 'sim_improved-learning.txt':
In the first couple of moves:
('right', 'red', None, 'right', 'left')
-- forward : 0.00
-- right : 0.53
-- None : 0.00
-- left : 0.00
The waypoint is to go right and the light is red. There is no oncoming traffic. The car coming from the left is turning right and the car coming from the right is turning left. The policy suggests that we should turn right, which is the optimal policy.
('forward', 'green', 'left', 'right', None) -- forward : 0.22 -- right : 0.63 -- None : 0.00 -- left : 0.00
In this example, the waypoint is forward, the light is green. There is oncoming traffic turning left (they should however give way to smartcab). Car from the left is turning right and there is no car coming from the right. The optimal policy is to go forward. Yet, the policy give above is to turn right, which is not in line with the optimal policy.
'gamma'Curiously, as part of the Q-Learning algorithm, you were asked to not use the discount factor, 'gamma' in the implementation. Including future rewards in the algorithm is used to aid in propogating positive rewards backwards from a future state to the current state. Essentially, if the driving agent is given the option to make several actions to arrive at different states, including future rewards will bias the agent towards states that could provide even more rewards. An example of this would be the driving agent moving towards a goal: With all actions and rewards equal, moving towards the goal would theoretically yield better rewards if there is an additional reward for reaching the goal. However, even though in this project, the driving agent is trying to reach a destination in the allotted time, including future rewards will not benefit the agent. In fact, if the agent were given many trials to learn, it could negatively affect Q-values!
There are two characteristics about the project that invalidate the use of future rewards in the Q-Learning algorithm. One characteristic has to do with the Smartcab itself, and the other has to do with the environment. Can you figure out what they are and why future rewards won't work for this project?
Answer:
The local environment is not static but is reset at each step. Moreover, the smartcab's final destination is not constant across trials.
Note: Once you have completed all of the code implementations and successfully answered each question above, you may finalize your work by exporting the iPython Notebook as an HTML document. You can do this by using the menu above and navigating to
File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.