ML Models and Tasks

 - Overview

Machine learning (ML) is a subset of artificial intelligence (AI). It enables systems to learn and improve themselves through experience without programming. It uses statistics to find patterns in large amounts of data, which can be many things like numbers, images or words, or whatever you have. 

Machine learning (ML) models are categorized into three main types: supervised, unsupervised, and reinforcement learning. 

Here are some ML tasks and model types: 

• Model deployment: One of the final steps in the ML lifecycle, this is the process of putting a model into real use to add value to an organization.
• Model retraining: A necessary stage because ML solutions are very data-dependent, and data trends change over time. Efficient retraining helps keep the solution relevant and saves the cost of recreating new solutions.
• Supervised learning: Machines are taught using examples, and a large amount of annotated data is used.
• Unsupervised learning: An ML algorithm analyzes and groups unlabeled data, finding hidden patterns and grouping them together based on similarities and differences.
• Data preparation: A crucial process that ensures ML models are built on high-quality data, leading to a more powerful model that's more accurate at making predictions.

- Machine Learning Models

A ML model can be understood as a program trained to find patterns in new data and make predictions. These models are expressed as a mathematical function that takes a request in the form of input data, makes a prediction on the input data, and then provides an output in response. 

First, these models are trained on a set of data, and then they are given an algorithm to reason about the data, extract patterns from the feed data, and learn from this data. Once these models are trained, they can be used to make predictions on unseen datasets.

There are various types of ML models that can be used depending on different business goals and datasets.

Some resources for running ML models include:

• Google Colaboratory
• Kaggle Kernels
• Paperspace
• Vast.ai
• Oracle Cloud
• Amazon Web Services (AWS)
• Microsoft Azure
• Google Cloud Platform (GCP)

- Walking through an Example of a ML Model

Let's walk through an example of a ML model, we are creating an application to recognize user's emotions based on facial expressions. Therefore, it is possible to create such an application through a ML model that we will train by feeding in images of faces labeled with various emotions. 

Whenever this application is used to determine the mood of a user, it reads all the data entered and then determines the mood of any user. So, in simple terms, we can say that a ML model is a simplified representation of some kind of thing or process. 

Some real-world examples of ML include: 

• Facial recognition
• Product recommendations
• Email automation and spam filtering
• Financial accuracy
• Social media optimization
• Healthcare advancement
• Mobile voice to text and predictive text 

- Four Categories of ML Methods

There are four types of ML: supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning. 

In short, in supervised learning, the input data is already defined (labeled) and you know what your output should be. In unsupervised learning, the input data is undefined (unlabeled) and you are not sure what your output is.

Literally, with ML, you show computer how to do certain things. For example, you want a computer to know how to cross a road. With conventional programming, you would give it a very precise set of rules, telling it how to look left and right, wait for cars, use pedestrian crossings, etc., and then let it go. 

With ML, you’d instead show it 10,000 videos of someone crossing the road safely (and 10,000 videos of someone getting hit by a car), and then let it do its thing. 

[Machine Learning Algorithms]

- Machine Learning Tasks

A ML task is a type of prediction or inference based on a question or problem posed and the data available. For example, classification tasks assign data to categories, and clustering tasks group data based on similarity.

ML tasks rely on patterns in data rather than being explicitly programmed. A ML task is a type of prediction or inference based on a question or problem posed and the data available. For example, classification tasks assign data to categories, and clustering tasks group data based on similarity. Machine learning tasks rely on patterns in data rather than being explicitly programmed.

Once you have determined which task is applicable to your scenario, you need to choose the best algorithm to train your model. 

Here are some ML tasks:

• Deep learning
• Reinforcement learning
• Cluster analysis
• Statistical classification
• Regression analysis
• Algorithm
• Dimensionality reduction
• Natural language processing
• Linear regression
• Anomaly detection
• Data analysis
• Feature extraction
• Multiclass classification
• Pattern recognition
• Exploratory data analysis
• Machine translation
• Feature engineering
• Hyperparameter optimization
• Forecasting
• Contextual image classification
• Image segmentation
• Analytics
• Data pre-processing
• Gradient descent
• Density estimation
• Data cleansing
• Object detection
• Software Testing
• Translation

- Probabilistic Models in Machine Learning

Probabilistic models are one of the most important parts of ML and are based on the application of statistical codes to data analysis. This goes back to one of the earliest ML methods and is still widely used today.

Unobserved variables are treated as random variables in a probabilistic model, and the interdependencies between variables are recorded in a joint probability distribution. It provides the basis for embracing the nature of learning. 

The probabilistic framework outlines a method for representing and deploying model reservations. In scientific data analysis, prediction plays a leading role. Their contributions are also critical in ML, cognitive computing, automation and AI. 

These probabilistic models have many admirable properties and are very useful in statistical analysis. They make it very simple to infer the inconsistencies present in most data. In fact, they can be built in layers to create complex models from basic elements.

One of the main reasons why probabilistic modeling is so popular today is that it provides natural protection against overfitting and allows for fully consistent inferences over complex forms of data.