You are on page 1of 9

NPTEL

Video Course on Machine Learning

Professor Carl Gustaf Jansson, KTH

Week 4 Inductive Learning based on


Symbolic Representations
and Weak Theories

Video 4.1 Inductive Learning based on Symbolic


Representations and Weak Theories
Symbolic Representations

Symbolic Representations Sub-symbolic Representations


Symbols, Lists Neural Networks
Semantic Networks Genetic Algorithms
Bayesian Networks Binary Strings
Production Rules, Logic
Decision Trees

Mathematics
Numbers, Sets, Graphs, Vectors, Matrices, Tensors
Weak Theories

Creating Abstractions from Creating and Modifying

sets of Instances with a Abstractions on the border


very weak or absent Model of an existing Model or
or Background Theory Background Theory

Normally a large set of A smaller set of instances


instances is required can be sufficiant
Lectures for this week

Generalization as Search

Decision Tree Learning Algorithms

Instance-based Learning

Clustering Analysis
Generalization as Search
Generalization as search in general
The hypothesis language
The ´more specific than´ relation
Partial Orderings
Generate and Test strategies for Generalization

Data Driven Strategies for Generalization

Depth first search


Breadth first search
Version Space Approach
Decision Tree Learning Algorithms

Decision Trees in general


TDIDT algorithms
Information theoretical measures
ID3 algoritm
Treating overfitting through pruning
Alternative algorithms
Instance-based Learning

Instance-based learning in general


The structure of the instance space
K-nearest neighbor algorithm
Distance and Similarity metrices
Weighted nearest neighbor algorithm
Binary Linear Classifier
Support Vector Machines
Kernel methods
The Kernel Trick enabling binary non-linear classification
Clustering Analysis

Clustering in general
Hyperparameters
Distance measures

Categories of clustering algorithms

Partitioning–based clustering
Hierarchical-based Clustering
Density-based clustering
Grid-based clustering
Model-based clustering
NPTEL

Video Course on Machine Learning

Professor Carl Gustaf Jansson, KTH

Thanks for your attention!

The next lecture 4.2 will be on the topic:

Generalization as Search

You might also like