APEX INSTITUTE OF TECHNOLOGY

COMPUTER SCIENCE &

ENGINEERING
Bachelor of Engineering (Computer Science)
Hypothesis Space & Inductive Bias
20CSF-286
Prof. (Dr.) Paras Chawla(E5653)

Unit 1 : Machine Learning DISCOVER . LEARN . EMPOWER

Outline
• Machine Learning
 Need of ML
 Learning Process
 ML Applications
• Types of ML :
Supervised Learning
Unsupervised Learning
Reinforcement Learning
• Hypothesis Space and Inductive Bias
 Hypothesis Representation
• A set of example days, and each is described by six attributes.
• The task is to learn to predict the value of Enjoy Sport for arbitrary day, based on the
values of its attribute values.

Enjoy Sport – Hypothesis Representation

• Each hypothesis will be a vector of six constraints, specifying the values of the six attributes – (Sky,
AirTemp, Humidity, Wind, Water, and Forecast).

Warm days
 Hypothesis Representation
• A hypothesis:
Sky AirTemp Humidity Wind Water Forecast
< Sunny, ? , ? , Strong , ? , Same >

• The most general hypothesis – that every day is a positive example

< ? , ?, ? , ? , ? , ? >

• The most specific hypothesis – that no day is a positive example

< 0 , 0, 0 , 0 , 0 , 0 >

• EnjoySport concept learning task requires learning the sets of days for which
EnjoySport=yes, describing this set by a conjunction of constraints over the instance
attributes.
Notation- EnjoySport Concept Learning Task
Hypothesis Space and Inductive Bias
• Hypothesis (h): A hypothesis is a function that best describes the target in supervised machine learning.  It is a
provisional idea that requires some evaluation like true and false.
• Hypothesis Space (H) : It is the set of all legal hypothesis. This is the set from which the machine learning
algorithm would determine the best possible (only one) which would best describe the target function or the
outputs.
h1 € H ( h1 : A hypothesis for Machine Learning Model )
• Good Hypothesis : A good hypothesis fits the evidence and can be used to make predictions about new
observations or new situations.
• Inductive bias: An inductive bias allows a learning algorithm to prioritize one solution (or interpretation) over
another, independent of the observed data by making two assumptions or biases :
1. Restrictions 2. Preferences
Hypothesis Space and Inductive Bias
• All these legal possible ways in which we can divide the coordinate plane to predict the outcome of the test
data composes of the Hypothesis Space.
• Each individual possible way is known as the hypothesis.
• Hence, in this example the hypothesis space would be like:
Inductive and Deductive Reasoning
Inductive Bias
Example
New Problem
Inductive Bias
Inductive Bias
THANK
YOU
For Queries,
Write at : drparaschawla.ece@cumail.in