Scribd
Upload
33 views

Machine Learning L1

Uploaded by

Akshaya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
33 views

Machine Learning L1

Uploaded by

Akshaya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 34

MACHINE LEARNING

Module 1 - Parameter Estimation and Regression


1. Overview of machine learning:
supervised, semi-supervised,
unsupervised learning,
reinforcement learning.
2. Basics of parameter estimation:
Maximum Likelihood
Estimation(MLE), Maximum a
Module 1 Posteriori Estimation (MAP).
3. Gradient Descent Algorithm,
Parameter Estimation and Batch Gradient Descent,
Stochastic Gradient Descent.
Regression 4. Regression algorithms: least
squares linear regression,
normal equations and closed
form solution, Polynomial
regression.
What is Machine Learning
Human can learn from past experiences and make decision on
its own
What is Machine Learning
What is Machine Learning
Just like, what we did to human, we need to provide experience
to the machine.
What is Machine Learning

“Learning is any process by which a system improves performance


from experience.”
- Herbert Simon

Definition by Tom Mitchell (1998):


Machine Learning is the study of algorithms that
● improve their performance P
● at some task T
● with experience E.
● A well-defined learning task is given by <P, T, E> .
When Do We Use Machine Learning?
ML is used when:

● Humans can’t explain their expertise (speech recognition)


● System needs to adapt to a changing environment (e.g. spam
detection)
● Models must be customized (personalized medicine).
● Models are based on huge amounts of data (genomics) want the
system to perform better than the human programmers.
When Do We Use Machine Learning?
Many domains and applications

Healthcare
❖ Medical Imaging: Used for detecting abnormalities
in X-rays, MRIs, and CT scans, such as identifying
tumors.
❖ Predictive healthcare analytics: ML models are
trained on medical data to predict diseases (e.g.,
cancer, diabetes) and patient outcomes.
Many domains and applications

Finance
❖ Algorithmic Trading: Predicts
market trends and automates
trading strategies based on large
datasets and real-time data.
❖ Credit Scoring: Assesses
creditworthiness by analyzing
behavioral and financial data.
Many domains and applications

Retail and E-Commerce


❖ Recommendation Systems: Personalized product
recommendations by analyzing user behavior and
preferences (e.g., Amazon, Netflix)
❖ Customer Service Automation: Chatbots and
virtual assistants use natural language processing
(NLP) to assist customers.
Many domains and applications

Autonomous Systems
❖ Self-driving Cars: ML is central to autonomous
driving systems, enabling vehicles to perceive the
environment and make driving decisions.
❖ Smart Homes: ML enables home automation
systems to learn user preferences and automate
household tasks (e.g., Nest thermostats).
Many domains and applications

Agriculture
❖ Crop Disease Detection: Identifies diseases in
plants and crops using image recognition.
❖ Automated Irrigation: Predicts optimal irrigation
schedules based on weather and soil conditions.
Types of Learning
Types of learning
● Supervised learning
○ Given: training data + desired outputs(labels)
● Unsupervised learning
○ Given: training data (without desired outputs)
● Semi-supervised learning
○ Given: training data + a fewdesired outputs
● Reinforcement learning
○ Rewards from sequence ofactions
Supervised learning

Given a labelled dataset, the task is to devise a function which takes


the dataset, and a new sample, and produces an output value.
Supervised learning

If the possible output values of the function are predefined and


discrete/categorical, it is called Classification
Supervised learning: Classification

Predefined classes means, it will produce output only from the labels
defined in the dataset. For example, even if we input a bus, it will
produce either CAR or BIKE
Supervised learning: Regression
Supervised learning

Supervised learning is the subcategory of machine


learning that focuses on learning a classification, or
regression model, that is, learning from labeled training
data.

Illustration of binary classification Illustration of linear regression


Unsupervised learning

In the unsupervised learning, we do not need to know the labels or Ground truth
values. The task is to identify the patterns like group the similar objects together.
Unsupervised learning
Unsupervised learning

unsupervised learning is a branch of machine learning


that is concerned with unlabeled data. Common tasks
in unsupervised learning are clustering analysis and
dimensionality reduction (compressing data onto a
lower-dimensional subspace or manifold).
Semi-supervised learning
● Semi-supervised learning can be described as a mix
between supervised and unsupervised learning.
● The model is trained on a small amount of labeled
data and a large amount of unlabeled data.
● The goal is to leverage the unlabeled data to improve
learning efficiency, especially when acquiring labeled
data is expensive or time-consuming.
Semi-supervised learning
● First train an unsupervised model on unlabeled data
● Then incorporate the model’s learned weights into a
supervised model and train it on the labeled data
Reinforcement learning
● Reinforcement is the process of learning from
rewards while performing a series of actions
Reinforcement learning
Reinforcement learning
Reinforcement learning
Reinforcement learning
● We do not tell the learner or agent which action to take
but merely assign a reward to each action and/or the
overall outcome.
● The learner must discover or learn a behavior that
maximizes the reward for a series of actions
● There is no such thing as the best action in any
intermediate state; an action is good if it is part of a good
policy.

You might also like