CodeNewbie Community

Cover image for Let's peek into machine learning
Dulya Perera
Dulya Perera

Posted on • Originally published at Medium

Let's peek into machine learning

Machine Learning and Artificial Intelligence are two of the most often used terms nowadays. Several real-world applications, notably Sophia, Arimac-Diyazen have already been built leveraging these ML and AI technologies.

What is Machine Learning?

Machine Learning(ML) is a subset of Artificial Intelligence(AI) that focuses on the creation of automated computer algorithms that require human intervention. As a result, humans must direct it toward a specific goal. We utilize computer power to examine previous examples in order to create a model that can predict the outcome of new examples.

For example, when Netflix offers a show or movie to you, it uses a model based on what you and other users have viewed to predict what you would prefer.

What are the variance of Machine Learning?

As new data is fed into machine learning algorithms, they learn and optimize their processes to increase performance, gradually acquiring intelligence. Machine Learning algorithms are classified into three types: Supervised Learning, Unsupervised Learning, and Reinforcement Learning.
ML

As indicated in the figure above, Supervised Learning is used to predict the output based on the acquired data, Unsupervised Learning is used to cluster the data or identify the hidden structure in data, and Reinforcement Learning is used to learn from a sequence of actions. Lets glance at each of them independently.

1. Supervised Learning
Source: Supervised Learning

Supervised Machine Learning is a learning method that uses labeled training data to predict outcomes for unlabeled data. Supervised learning involves training the system using well-labeled data. It signifies that some material has already been labeled with the correct responses.

Supervised learning algorithms are good for the following tasks:

  • Regression modeling: Predicting continuous values.
  • Ensembling: Combining the predictions of multiple machine learning models to produce an accurate prediction.

2. Unsupervised Learning
Source: Unsupervised Learning

Unsupervised Learning is a machine learning technique in which the machine/system will get trained by the information that is neither classified nor labeled and accept the output without guidance. Without prior data training, the end outcome is to categorize the unsorted information based on similarities or patterns. As shown in above Figure, the unlabeled data are processing using the algorithm and then provided the categorized information.

Unsupervised learning algorithms are good for the following tasks:

  • Clustering: Splitting the dataset into groups based on similarity.
  • Anomaly detection: Identifying unusual data points in a data set.
  • Association mining: Identifying sets of items in a data set that frequently occur together.
  • Dimensionality reduction: Reducing the number of variables in a data set.

3. Reinforcement Learning
Source: Reinforcement Learning

Reinforcement learning is a machine learning approach where the system gets the rewards or points for taking suitable action in a particular situation. The rewarding method of the machine is that for each correct step / action / behavior taken by the machine or system will give the reward while each wrong step / action / behavior will subtract the reward of the machine.

Reinforcement learning is often used in areas such as:

  • Robotics: Robots can learn to perform tasks the physical world using this technique.
  • Video gameplay: Reinforcement learning has been used to teach bots to play a number of video games.
  • Resource management: Given finite resources and a defined goal, reinforcement learning can help enterprises plan out how to allocate resources.

Consider a sweeper robot that cleans up the unclean spots in a room. The purpose of the sweeper robot, according to Reinforcement Learning, is to collect the reward by discovering the dirty spots and cleaning them while avoiding the other cleaned up areas. As a result, the sweeper robots seek to find all possible ways or paths to the reward while causing the least amount of disruption. By taking action and learning via trial and error, the robot can conceive and comprehend the surroundings.


Conclusion

Machine Learning is a minor component of Artificial Intelligence when compared to other types of AI. There is a General Artificial Intelligence in which a system can achieve everything without the intervention of a human. Nevertheless, when considering Artificial Intelligence, this is more where we could all headed.

Confused

The trick is to find the unbaked ones.

Discussion (0)