1. NADAR SARASWATHI COLLEGE OF
ARTS AND SCIENCE
DEPARTMENT OF COMPUTER SCIENCE
TOPIC:ENSEMBLE METHODS
Presented by..,
M.Shakthi Msc(cs)
2. ENSEMBLE METHODS:
Ensemble methods are techniques that create multiple models and then
combine them to produce improved results.
Ensemble methods in machine learning usually produce more accurate
solutions than a single model would.
Ensemble learning helps improve machine learning results by combining
several models. This approach allows the production of better predictive
performance compared to a single model.
An ensemble method is a technique which uses multiple independent
similar or different models/weak learners to derive an output or make some
predictions.
For e.g. A random forest is an ensemble of multiple decision trees.
3. THREE ENSEMBLE METHODS:
1.BOOSTING
2.BAGGING
3.STACKING
The most popular ensemble methods are boosting, bagging, and
stacking.
Ensemble methods are ideal for regression and classification, where they
reduce bias and variance to boost the accuracy of models.
4. THREE TYPES OF EMSEMBLE:
Three types of ensembles that is,
1. Micro canonical,
2. Canonical and
3.grand Canonical
MICRO CANONICAL:
The microcanonical assemble is a collection of essentially
independent assemblies having the same energy E, volume V and number of systems
N. The individual systems of a microcanonical ensemble are separated by rigid
impermeable and we'll insulated walls such that the values of E, V and N for a
particular system are not affected by the presence of other system.
5. CANONICAL:
The Canonical ensemble is a collection of essentially independent
assemblies having the same temperature T volume V and number of identical
particles N. The disparate systems of a canonical ensemble are separated by
rigid, impermeable but conducting walls.
Thus in canonical ensemble can exchange energy but not particles. The
quality of temperature of all the systems can be achieved by a bearing each in
thermal contact with the large heat reservoir at constant temperature T.
6. STACKING:
It is the collection of a large number of essentially independent
systems having the same temperature T, volume V and chemical potential (μ).
The individual system of grand canonical ensemble are separated by
rigid, permeable and conducting walls.
As the separating walls are conducting and permeable, the exchange
of heat energy as well as that of particles between the systems takes place in
such a way that all the system arrive at common temperature T and chemical
potential (μ).
7. WHY EMSEMBLE MODEL USED:
Ensembles are used to achieve better predictive performance on a
predictive modeling problem than a single predictive model.
The way this is achieved can be understood as the model reducing the
variance component of the prediction error by adding bias
(i.e. in the context of the bias-variance trade-off).
Advantages of ensemble methods:
1. Ensemble methods have higher predictive accuracy, compared to the
individual models.
2. Ensemble methods are very useful when there is both linear and non-
linear type of data in the dataset; different models can be combined to
handle this type of data.
8.
9. Disadvantage of ensemble method:
Ensembling is less interpretable, the output of the ensembled model is hard
to predict and explain. ...
The art of ensembling is hard to learn and any wrong selection can lead to
lower predictive accuracy than an individual model.
Ensembling is expensive in terms of both time and space.