Model Fitting: Overfitting, Underfitting, and Balanced

Understanding model fitting is important for understanding the models’ poor accuracy.

https://docs.aws.amazon.com/machine-learning/latest/dg/model-fit-underfitting-vs-overfitting.html

Overfitting: When the model performs too well on training data then it reduces the model flexibility for new data.

Underfitting: When the model performs poorly on the training data. It’s often caused by an excessively simple model.

Both overfitting and underfitting lead to poor performance in real time.

Balanced: Bbalanced models would show better accuracy on new data.



Machine Learning 2: Machine Learning Solutions

There are two ways machine intelligence can be part of your applications.

As a Service:

Machine intelligence requires three ingredients: computing power, algorithms, and data. If you’re new to business, i would suggest not to invest in researching these topics or building a team, instead focus on application development using machine intelligence API from companies like Microsoft, IBM, Google, Amazon, and others.

Example: Azure Cognitive Services, Amazon Translate, Amazon Polly,
Amazon Lex, etc.

Find the detailed comparison for Machine Learning as a Service here.

Custom Solution:

Real business problems are complex and solutions for those can’t be achieved using machine intelligence as a service. In this case, build your own machine learning solutions. Though machine learning frameworks and sdks are available in almost all programming language, But I’d recommend python, JavaScript, and .Net due the support from open source communities. These are some important platforms where you can deploy your solutions:

Azure Machine Learning and Azure ML Studio

Amazon SageMaker

IBM Watson Machine Learning

Google Cloud Machine Learning

Artificial Intelligence vs Machine Learning vs Deep Learning

Artificial Intelligence:

A branch of computer science dealing with the simulation of intelligent behavior in computers using human behavior such as visual perception, speech recognition, decision-making, and translation between languages. .

Machine Learning:

Basic approach of Machine learning is solving problems using algorithms in three steps “parse data”, “learn from it”, and “make a determination or prediction about something”.

Deep Learning:

Deep learning is a sub-set of machine learning with better accuracy. Deep learning structures algorithms in layers to create an “artificial neural network” that can learn and make intelligent decisions on its own.

Resources:

https://skymind.ai/wiki/ai-vs-machine-learning-vs-deep-learning

https://blogs.nvidia.com/blog/2016/07/29/whats-difference-artificial-intelligence-machine-learning-deep-learning-ai/

Machine Learning 1: What is Machine Learning? | Types of Machine Learning?

What is Machine Learning?

Machine Learning is the branch of computer science for developing algorithms and statistical models which computer systems can use to solve specific type of tasks.

Mathematics for Machine Learning:

  • Algebra (Equations, Vectors, Matrices)
  • Calculus(Function, Derivatives, Integrals)
  • Graphs
  • Statistics
  • Probability
  • Set Theory

Types of Machine Learning?

All machine learning algorithms are categorized into three categories:

  • Supervised and semi-supervised learning
  • Unsupervised leaning
  • Reinforcement learning

Create a website or blog at WordPress.com

Up ↑