Architectural Blueprints—The “4+1” View Model of Machine Learning

Architectural Blueprints—The “4+1” View Model of Machine Learning

Architectural Blueprints—The “4+1” View Model of Machine Learning

Based on the inspiration from “Architectural Blueprints—The “4+1” View Model of Software Architecture" [1], which “presents a model for describing the architecture of software-intensive systems, based on the use of multiple, concurrent views,” we developed “Architectural Blueprints—The “4+1” View Model of Machine Learning (ML).” We can capture it in this formula: ???

ML Architectural Blueprints = {Scenarios, Accuracy, Complexity, Interpretability, Operations}


ML is a form of Artificial Intelligence (AI), which makes predictions and decisions from data. It is the result of training algorithms and statistical models to analyze and draw inferences from patterns in data, which are able to learn and adapt without following explicit instructions. However, you need to:

  1. Check - double check your Assumptions,
  2. Mitigate - make sure you mitigate your Biases, and
  3. Validate - take your time to validate your Constraints.

The Assumptions, Biases, and Constraints (ABC) of data science, Data, and Models of ML can be captured in this formula: ???

Machine Learning = {Assumptions/Biases/Constraints, Data, Models}


Architectural Blueprints—The “4+1” View Model of Machine Learning

Architectural Blueprints—The “4+1” View Model of Machine Learning

ML Architectural Model:

To capture a machine learning architecture, we use a “4+1” ML view model. This model is composed of five main views:

  • The Accuracy view, which is a degree measure of how reliable is the conclusion,
  • the Complexity view, which is approximated for a variety of different machine learning algorithms during training and prediction for Time, Space, and Sample,
  • the Interpretability view, which measures how well the model enables understanding of the justification and reasoning to the decision conclusion,
  • the Operations view, which are the ML Operations (MLOps) of Continuous Integration (CI), Continuous Delivery/Deployment (CD), Continuous Training (CT), and Continuous Monitoring (CM).


Architectural Blueprints of ML: Scenarios, Accuracy, Complexity, Interpretability, & Operations Views

Architectural Blueprints of Machine Learning: Scenarios, Accuracy, Complexity, Interpretability, and Operations Views


ML Architectural Blueprints = {Scenarios, Accuracy, Complexity, Interpretability, Operations}


  • Scenarios

Your machine learning scenarios depend on learning categories, data types, and objectives. The four major categories are supervised, semi-supervised, unsupervised, and reinforcement. Data types include discrete, continuous, qualitative, quantitative, or large data items. Your objectives can be to predicate a category or a number, divide by similarity, or identify sequences.?

Read the "Scenarios: Which Machine Learning (ML) to choose?" [2] article at https://www.dhirubhai.net/pulse/machine-learning-101-which-ml-choose-yair-rajwan-ms-dsc.


  • ?Accuracy View

"Model fitting is a measure of how well [optimize] a machine learning model generalizes to similar [evaluation] data to that on which it was trained. A model that is well-fitted [optimal-fitted] produces more accurate outcomes. A model that is overfitted matches the data too closely. A model that is underfitted does not match closely enough."

Accuracy is affected by your model fitting. And, model fitting depends on Bias-Variance Trade-off in machine learning. Balancing bias and variance can solve overfitting and underfitting.

Read the "Accuracy: The Bias-Variance Trade-off" [3] article at https://www.dhirubhai.net/pulse/accuracy-bias-variance-tradeoff-yair-rajwan-ms-dsc.


  • Complexity View

Computational complexity of an algorithm is a fundamental concept in computer science. It is necessary to be taken into account because it affects the amount of resources required to run your model. These required resources are estimated for time, space, and sample complexities.

These estimated computational complexities are approximated for a variety of different machine learning algorithms during Training and Prediction.

Read the "Complexity: Time, Space, & Sample" [4] article at https://www.dhirubhai.net/pulse/complexity-time-space-sample-yair-rajwan-ms-dsc.


  • Interpretability View

In order to understand and trust your model prediction/decision and reasoning, there are two tantamount and equivalent factors that need to be considered: Interpretability and Explainability.

Interpretability is “The ability to determine and observe cause and effect from a machine learning.”

Explainability is “The ability to justify the reason and its importance of a machine learning results.”

Read the “Interpretability: "Seeing Machines Learn” – Understanding ML models' prediction/decision" [5] article at https://www.dhirubhai.net/pulse/interpretabilityexplainability-understanding-models-rajwan-ms-dsc. ?


  • Operations View

In machine learning, ML Operations (MLOps) and Continuous Machine Learning (CML) are a set of practices that aims to deploy and maintain machine learning models in production reliably and efficiently. MLOps includes techniques and tools for implementing and automating ML pipelines: Continuous Integration (CI), Continuous Delivery/Deployment (CD), Continuous Training/Testing (CT), and Continuous Monitoring (CM) - MLOps (CI/CD/CT/CM).

Read the "Operations: ML Operations (MLOps), Continuous ML & AutoML" [6] article at https://www.dhirubhai.net/pulse/ml-operations-mlops-continuous-automl-yair-rajwan-ms-dsc.


Next, read the "Data Science Approaches to Data Quality: From Raw Data to Datasets" article at?https://www.dhirubhai.net/pulse/data-science-approaches-quality-from-raw-dataset-yair-rajwan-ms-dsc.

---------------------------------------------------------?

[1] https://ieeexplore.ieee.org/document/469759

[2] https://www.dhirubhai.net/pulse/machine-learning-101-which-ml-choose-yair-rajwan-ms-dsc

[3] https://www.dhirubhai.net/pulse/accuracy-bias-variance-tradeoff-yair-rajwan-ms-dsc

[4] https://www.dhirubhai.net/pulse/complexity-time-space-sample-yair-rajwan-ms-dsc

[5] https://www.dhirubhai.net/pulse/interpretabilityexplainability-understanding-models-rajwan-ms-dsc

[6] https://www.dhirubhai.net/pulse/ml-operations-mlops-continuous-automl-yair-rajwan-ms-dsc

Next, read the "Data Science Approaches to Data Quality: From Raw Data to Datasets" article at?https://www.dhirubhai.net/pulse/data-science-approaches-quality-from-raw-dataset-yair-rajwan-ms-dsc

回复

Read the "Machine Learning 101 – Which Machine Learning (ML) to choose?" article at?https://www.dhirubhai.net/pulse/machine-learning-101-which-ml-choose-yair-rajwan-ms-dsc

回复

要查看或添加评论,请登录

社区洞察

其他会员也浏览了