O'Reilly logo

Autonomous Learning Systems: From Data Streams to Knowledge in Real-time by Plamen Angelov

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

7

Autonomous Predictors, Estimators, Filters, Inferential Sensors

The autonomous learning systems (ALS) concept described in this book is quite generic and can be applied to numerous problems. They can be summarised as:

A. clustering (unsupervised learning, multiple inputs, no output, MINO);
B. predictors, estimators, filters, inferential sensors (semisupervised learning, multiple inputs, multiple outputs, MIMO);
C. classifiers (semisupervised learning or unsupervised learning, MISO for the so-called two-class problem and MIMO for the general multiclass classification problem);
D. controllers (semisupervised learning; usually MISO, but can be MIMO).

Clustering was described in Section 3.2. In the context of ALS one can use AutoCluster or the ELM approach; that is, evolving clustering methods, which were described in Section 3.2.3.

7.1 Predictors, Estimators, Filters – Problem Formulation

In this chapter the problem B as itemised above will be described, namely, predictors, estimators, filters and inferential sensors. These seemingly different problems that are subject to various disciplines such as forecasting and statistical learning, signal processing, chemical industry automation, system identification, and so on can be combined in the simplistic representation of Figure 7.1.

This extremely simplistic diagram represents a vector of inputs, being transformed into the outputs, ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required