You work as a machine learning specialist for a robotics product manufacturer.
Your company is trying to use machine learning to help its automatic vacuuming robot determine the most efficient path across the floor of a room.
You need to build a machine learning model to accomplish this problem. Which modeling approach best fits your problem?
Click on the arrows to vote for the correct answer
A. B. C. D.Answer: B.
Option A is incorrect.
Multi-Class Classification is used when your model needs to has many class outcomes from which to choose, as in a car model classification image recognition problem.
In this strategy determination problem, we need to learn a strategy that optimizes an objective.
A Multi-Class Classification approach wouldn't give you this result.
Option B is correct.
Simulation-Based Reinforcement Learning is used in problems where your model needs to learn through trial and error.
This is how a robot would best learn the optimal path through a given environment.
Option C is incorrect.
Binary Classification is the approach you use when you are trying to predict a binary outcome.
This strategy determination problem would not fit a binary classification model.
Option D is incorrect.
The Heuristic Approach is used when a machine learning approach is not necessary.
An example is the rate of acceleration of a particle through space.
There are well known formulas for speed, inertia, and friction that can solve a problem such as this.
Reference:
Please see the Amazon SageMaker developer guide titled Linear Learner Algorithm, the Amazon SageMaker developer guide titled Reinforcement Learningwith Amazon SageMaker RL,the Amazon Machine Learning developer guide titled Multiclass Classification, and the article titled What is the difference between a machine learning algorithm and a heuristic, and when to use each?
The best modeling approach for this problem is B. Simulation-based Reinforcement Learning.
Reinforcement Learning is a branch of machine learning that involves an agent interacting with an environment to learn how to take actions that maximize a reward. In this case, the environment is the room, and the agent is the automatic vacuuming robot. The goal of the agent is to learn the most efficient path across the floor of the room, which would maximize its cleaning performance while minimizing the time and energy required.
Simulation-based Reinforcement Learning involves training an agent in a simulated environment, which is a digital representation of the physical environment. The agent can then learn by taking actions and receiving feedback in the simulated environment without causing any damage or harm to the physical environment. Once the agent is trained, it can be deployed in the physical environment to navigate the vacuuming robot across the room.
The other modeling approaches listed are not as suitable for this problem:
A. Multi-Class Classification: This approach is used when the task is to classify instances into one of several classes. However, in this problem, the task is not to classify instances into classes but to find the most efficient path across the floor of the room.
C. Binary Classification: This approach is used when the task is to classify instances into one of two classes. However, in this problem, the task is not to classify instances into classes but to find the most efficient path across the floor of the room.
D. Heuristic Approach: A heuristic approach involves using a rule-based system to solve a problem. However, in this problem, the most efficient path may not be obvious and may require the robot to explore and learn through experience, which is better suited to a machine learning approach.