Comments about "Machine learning" in Wikipedia

This document contains comments about the article Machine learning in Wikipedia
In the last paragraph I explain my own opinion.

Contents

Reflection

Introduction

The article starts with the following sentence.

1. Overview

2. History and relationships to other fields

2.1 Artificial intelligence

2.2 Data mining

2.3 Optimization

2.4 Generalization

2.5 Statistics

2.6 Physics

3.Theory

4.Approaches

4.1 Supervised learning

4.2 Unsupervised learning

4.3 Semi-supervised learning

4.4 Reinforcement learning

4.5 Dimensionality reduction

4.6 Other types

4.6.1 Self-learning

4.6.2 Feature learning

4.6.3 Sparse dictionary learning

4.6.4 Anomaly detection

4.6.5 Robot learning

4.6.6 Association rules

5 Models

5.1 Artificial neural networks

5.2 Decision trees

5.3 Support-vector machines

5.4 Regression analysis

5.5 Bayesian networks

5.6 Gaussian processes

5.7 Genetic algorithms

5.8 Training models

5.8.1 Federated learning

6. Applications

7.Limitations

Although machine learning has been transformative in some fields, machine-learning programs often fail to deliver expected results.
Always what you expect that a machine should learn, a human should also be able to perform.
For example: if a doctor can not heal a patient, you should not expect that a machine-learning program can
Reasons for this are numerous: lack of (suitable) data, lack of access to the data, data bias, privacy problems, badly chosen tasks and algorithms, wrong tools and people, lack of resources, and evaluation problems
For example: If the machine is not trained to do an operation, than you cann't expect the machine does. All the other reasons are more or less the same.

7.1 Explainability

Explainable AI (XAI), or Interpretable AI, or Explainable Machine Learning (XML), is artificial intelligence (AI) in which humans can understand the decisions or predictions made by the AI.
But they must also be the best. That means they should be agree with a human test case, using the same data base.
It contrasts with the "black box" concept in machine learning where even its designers cannot explain why an AI arrived at a specific decision.
Such an answer is considered wrong. The system should also explain why it comes to such a 'wrong' conclusion. . If required should show all the intermediate steps.
XAI may be an implementation of the social right to explanation.
AI should always have a mode that it shows all the intermediate steps

7.2 Overfitting

Settling on a bad, overly complex theory gerrymandered to fit all the past training data is known as overfitting.
This sentence is not clear.

7.3 Model collapse

7.4 Hallucinations

7.5 Other limitations and vulnerabilities

8.Model assessments

9.Ethics

10.Hardware

10.1 Neuromorphic/Physical Neural Networks

10.2 Embedded Machine Learning

11.Software

11.1 Free and open-source software

11.2 Proprietary software with free and open-source editions

11.3 Proprietary software

12.Journals

13.Conferences

14. See also

Following is a list with "Comments in Wikipedia" about related subjects
Unsolved Problems in ML Safety
https://arxiv.org/pdf/2109.13916

Reflection 1

Reflection 2

Reflection 3

Feedback

If you want to give a comment you can use the following form Comment form
Created: 11 January 2022

Go Back to Wikipedia Comments in Wikipedia documents
Back to my home page Index