31-08-2025

Machine Learning
• In the real world, we are surrounded by humans
who can learn everything from their experiences
with their learning capability, and we have
computers or machines which work on our
instructions.
Introduction to Machine Learning • But can a machine also learn from experiences or
past data like a human does? So here comes the
role of Machine Learning.
• Machine Learning is said as a subset of artificial
intelligence that is mainly concerned with the
development of algorithms which allow a computer
to learn from the data and past experiences on
their own.

Artificial intelligence
Machine Learning
• A Machine Learning system learns from historical data, builds the prediction models, and • Concisely, AI can be described as the effort
whenever it receives new data, predicts the output for it. to automate intellectual tasks normally
• The accuracy of predicted output depends upon the amount of data, as the huge amount of data performed by humans.
helps to build a better model which predicts the output more accurately. • AI is a general field that encompasses
• Suppose we have a complex problem, where we need to perform some predictions machine learning and deep learning
• Instead of writing a code for it, we need to feed the data to generic algorithms, and with the help • AI also includes many more approaches that
of these algorithms, machine builds the logic as per the data and predict the output. may not involve any learning.

Deep Learning with Python – Francois Chollet

1
 31-08-2025

Artificial intelligence Artificial intelligence

• Symbolic AI: Dominant paradigm in AI from the 1950s -1980
• Reached peak popularity during the expert systems boom of the 1980s • Symbolic AI to Machine Learning
• Early chess programs, only involved hardcoded rules crafted by programmers and did not qualify as • Symbolic AI proved suitable to solve well-defined, logical problem
machine learning. • such as playing chess
• For a fairly long time, most experts believed that human-level artificial intelligence could be achieved • Turned out to be intractable to figure out explicit rules for solving more complex, fuzzy problem
by having programmers handcraft a sufficiently large set of explicit rules for manipulating knowledge • image classification, speech recognition, or natural language translation.
stored in explicit databases.
• A new approach arose to take symbolic AI’s place: machine learning
• This approach is known as symbolic AI.

Note: The origins of chess are debated, but most historians agree that the game's ancestor, chaturanga, originated
in India around the 6th century AD. Chaturanga is a Sanskrit word that refers to a type of battle formation, and is
mentioned in the Mahabharatha.

Deep Learning with Python – Francois Chollet Deep Learning with Python – Francois Chollet

Machine Learning Machine Learning: a new programming paradigm

• Machine learning:
• Machine looks at the input data and the corresponding answers, and figures out what the rules
should be
• A machine learning system is trained rather than explicitly programmed.
• It is presented with many examples relevant to a task, and it finds statistical structure in these
examples that eventually allows the system to come up with rules for automating the task.

Deep Learning with Python – Francois Chollet Deep Learning with Python – Francois Chollet

2
 31-08-2025

Machine Learning Machine Learning

• Unlike statistics, machine learning tends to deal with large, complex datasets (such as a dataset of To do machine learning, we need three things:
millions of images, each consisting of tens of thousands of pixels)
1. Input data points—
• Classical statistical analysis such as Bayesian analysis would be impractical.
• For instance, if the task is speech recognition, these data points could be sound files of
• Machine learning, and especially deep learning, exhibits comparatively little mathematical theory people speaking.
and is fundamentally an engineering discipline • If the task is image tagging, they could be pictures.
2. Examples of the expected output—
• In a speech-recognition task, these could be human-generated transcripts of sound files.
• In an image task, expected outputs could be tags such as “dog,” “cat,” and so on.
3. A way to measure whether the algorithm is doing a good job—
• This is necessary to determine the distance between the algorithm’s current output and its
expected output.
• The measurement is used as a feedback signal to adjust the way the algorithm works.
• This adjustment step is what we call learning.
Deep Learning with Python – Francois Chollet Deep Learning with Python – Francois Chollet

Motivation Applications and type of data [Link]

Automatic translation of documents
• Over the past two decades Machine Learning has become one of the mainstays of information • At one extreme, we could aim at fully understanding a text before translating it
technology
• using a curated set of rules crafted by a computational linguist well versed in the two languages
• With the ever increasing amounts of data becoming available, there is good reason to believe that • Arduous task
smart data analysis will become more pervasive as a necessary ingredient for technological
• text is not always grammatically correct,
progress
• document understanding is not a trivial task
• Instead, we could use examples of translated documents, such as multilingual entities (United
Nations, European Union, Switzerland) to learn how to translate between the two languages
• We could use examples of translations to learn how to translate.
• This machine learning approach proved quite successful
BHASHINI aims to transcend language barriers, ensuring that every citizen can effortlessly access digital services in
their own language. Using voice as a medium, BHASHINI has the potential to bridge language as well as the digital
divide. Launched by Honourable PM Shri Narendra Modi in July 2022 under the National Language Technology
Mission, BHASHINI aims to provide technology translation services in 22 scheduled Indian languages.
Introduction to Machine Learning Alex Smola and S.V.N. Vishwanathan

3
 31-08-2025

Introduction Introduction
• The term machine learning refers to the automated detection of meaningful patterns in data. Common feature of all of these applications
• In the past couple of decades it has become a common tool in almost any task that requires • In contrast to more traditional uses of computers, due to the complexity of the patterns that need
information extraction from large data sets. to be detected, a human programmer cannot provide an explicit, fine detailed specification of
how such tasks should be executed.
• We are surrounded by a machine learning based technology:
Search engines learn how to bring us the best results (while placing profitable ads), • Ex: we human beings, many of our skills are acquired or refined through learning from our
experience (rather than following explicit instructions given to us)
Anti-spam software learns to filter our email messages, and
Credit card transactions are secured by a software that learns how to detect frauds. • Machine learning tools are concerned with endowing programs with the ability to learn and
adapt.
Digital cameras learn to detect faces and
Intelligent personal assistance applications on smart-phones learn to recognize voice commands
Cars are equipped with accident prevention systems that are built using machine learning algorithms.
Machine learning is also widely used in scientific applications such as bioinformatics, medicine, and astronomy

Understanding Machine Learning: From Theory to Algorithms by Shai Shalev-Shwartz and Shai Ben-David

When Do We Need Machine Learning?

Introduction
• Automated learning, we more often call, Machine Learning (ML) • When do we need machine learning rather than directly program our computers to carry out the task
• That is, we wish to program computers so that they can “learn" from input available to at hand
them. • Two aspects of a given problem may call for the use of programs that learn and improve on the basis
• Roughly speaking, learning is the process of converting experience into expertise or of their “experience":
knowledge. 1. The problem's complexity and
• The input to a learning algorithm is training data, representing experience, and 2. the need for adaptivity
• the output is some expertise, which usually takes the form of another computer program that can
perform some task.
• Seeking a formal-mathematical understanding of this concept, we'll have to be more
explicit about what we mean by each of the involved terms:
• What is the training data our programs will access?
• How can the process of learning be automated?
• How can we evaluate the success of such a process (namely, the quality of the output of a learning
program)?

4
 31-08-2025

When Do We Need Machine Learning? When Do We Need Machine Learning?

1. Tasks That Are Too Complex to Program 1. Tasks That Are Too Complex to Program

A. Tasks Performed by Humans: There are numerous tasks that we perform routinely B. Tasks beyond Human Capabilities:
• yet we cannot sufficiently elaborate to extract a well defined program. • Another wide family of tasks that benefit from machine learning techniques are related to the analysis
• Ex: driving, speech recognition, and image understanding. of very large and complex data sets:
• astronomical data, turning medical archives into medical knowledge, weather prediction, analysis
• In all of these tasks, state of the art machine learning programs, programs that “learn from their of genomic data, Web search engines, and electronic commerce.
experience," achieve quite satisfactory results, once exposed to sufficiently many training
examples. • With more and more available digitally recorded data, it becomes obvious that there are treasures of
meaningful information buried in data archives that are way too large and too complex for humans to
make sense of.
• Learning to detect meaningful patterns in large and complex data sets is a promising domain in which
the combination of programs that learn with the almost unlimited memory capacity and ever increasing
processing speed of computers opens up new horizons.

When Do We Need Machine Learning?

Relation to other fields
2. Adaptivity
• As an interdisciplinary field machine learning shares common threads
• One limiting feature of programmed tools is their rigidity - once the program has been written down
and installed, it stays unchanged. • With mathematical fields of statistics, information theory, game theory, and optimization
• However, many tasks change over time or from one user to another. • It is naturally a subfield of computer science
• Machine learning tools - programs whose behavior adapts to their input data - offer a solution to such • as our goal is to program machines so that they will learn.
issues; • Machine learning can be viewed as a branch of AI (Artificial Intelligence),
• They are, by nature, adaptive to changes in the environment they interact with. • the ability to turn experience into expertise or to detect meaningful patterns in complex sensory
data is a cornerstone of human intelligence
• Typical successful applications of machine learning to such problems include
• programs that decode handwritten text, where a fixed program can adapt to variations between
the handwriting of different users;
• spam detection programs, adapting automatically to changes in the nature of spam e-mails; and
• speech recognition programs

5
 31-08-2025

Relation to other fields Relation to other fields

• ML and Statistics ML and Statistics: Similarities
 The component of experience, or training, in machine learning often refers to data that is • ML and Statistics: There is a lot in common between the two disciplines, in terms of both the goals and
randomly generated. techniques used.
 The task of the learner is to process such randomly generated examples toward drawing • if a doctor comes up with the hypothesis that there is a correlation between smoking and heart disease
conclusions that hold for the environment from which these examples are picked. • it is the statistician's role to view samples of patients and check the validity of that hypothesis (this
 This description of machine learning highlights its close relationship with statistics is the common statistical task of hypothesis testing).
• In contrast, machine learning aims to use the data gathered from samples of patients to come up with a
description of the causes of heart disease.
• The hope is that automated techniques may be able to figure out meaningful patterns (or
hypotheses) that may have been missed by the human observer.

Relation to other fields Relation to other fields

• ML vs Statistics: Differences ML vs Statistics: Differences
• In contrast with traditional statistics, in machine learning algorithmic considerations play a major role. 3. In statistics it is common to work under the assumption of certain presubscribed data models (such as
• Machine learning is about the execution of learning by computers; assuming the normality of data-generating distributions, or the linearity of functional dependencies)
• hence algorithmic issues are pivotal. • in machine learning the emphasis is on working under a “distribution-free" setting, where the learner
assumes as little as possible about the nature of the data distribution and allows the learning algorithm
1. We develop algorithms to perform the learning tasks and are concerned with their computational to figure out which models best approximate the data-generating process.
efficiency.
2. Another difference is that while statistics is often interested in asymptotic behavior (like the
convergence of sample-based statistical estimates as the sample sizes grow to infinity),
• theory of machine learning focuses on finite sample bounds.
• Namely, given the size of available samples, machine learning theory aims to figure out the degree of
accuracy that a learner can expect on the basis of such samples.

6
 31-08-2025

What Is Machine Learning? – Informal Definition

First ML Applications? • Machine learning can be defined as the process of solving a practical problem by
• ML has been around in some specialized applications, such as Optical Character Recognition (OCR).
• First ML application that really became mainstream in in the 1990s: the spam filter. 1) gathering a dataset, and
• It does technically qualify as Machine Learning (actually learned so well that we seldom need to flag an 2) algorithmically building a statistical model based on that dataset
email as spam anymore).
• It was followed by hundreds of ML applications that now quietly power hundreds of products and
features that we use regularly • That statistical model is assumed to be used to solve the practical problem.
• from better recommendations to voice search

Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron The Hundred-Page Machine Learning Book Andriy Burkov

What Is Machine Learning? – Formal Definition

WELL-POSED MACHINE LEARNING PROBLEMS
• Machine Learning is the science (and art) of programming computers so they can learn from data.
In general, to have a well-defined learning problem, we must identify these three features:
• Definitions: • The learning task
• [Machine Learning is the] field of study that gives computers the ability to learn without being explicitly • The measure of performance
programmed. —Arthur Samuel, 1959 • The task experience
• Engineering-oriented: A computer program is said to learn from experience E with respect to some task • The computer program is the ‘machine’ in our context.
T and some performance measure P, if its performance on T, as measured by P, improves with
experience E. —Tom Mitchell, 1997 • The computer program is designed employing learning from the task experience.
• Equivalently, we say that the machine is trained using task experience, or machine learns from task
experience
• Understanding the inputs and outputs is of greater importance

Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron Applied Machine Learning - Gopal

7
 31-08-2025

WELL-POSED MACHINE LEARNING PROBLEMS WELL-POSED MACHINE LEARNING PROBLEMS

• The input is defined by the learning task. • Numerical form of data representation allows us to deal with patterns geometrically
• So we use learning algorithms using linear algebra and analytic geometry
• Four different types of learning tasks appear in the real-world applications (supervised, unsupervised,
reinforcement, learning based on natural processes) • Characterizing the similarity of patterns in state space can be done through some form of
• metric (distance) measure:
• For different forms of raw data (text, images, waveforms, and so forth), it is common to represent data
in standard fixed length vector formats with numerical values. • distance between two vectors is a measure of similarity between two corresponding patterns.
• Many measures of ‘distance’ have been proposed in the literature.
• Such abstractions typically involve significant loss of information, yet they are essential for a well-
defined learning problem. • In another class of machine leaning problems, the input (experience) is available in the form of
• nominal (or categorical) data, described in linguistic form (not numerical).
• For nominal form of data, there is no natural notion of similarity.
• Each learning algorithm based on nominal data employs some nonmetric method of similarity.

Applied Machine Learning - Gopal Applied Machine Learning - Gopal

WELL-POSED MACHINE LEARNING PROBLEMS WELL-POSED MACHINE LEARNING PROBLEMS

• The output of an algorithm represents the learned knowledge. A block diagrammatic representation of a learning machine
• This knowledge is in the form of a model of the structural patterns in the data.
• The model is deployed by the user for decision-making;
• it gives the prediction with respect to the assigned task for measurements/observations not in the task
experience;
• a good model will generalize well to observations unseen by the machine during training.

Applied Machine Learning - Gopal Applied Machine Learning - Gopal

8
 31-08-2025

Spam filter Application using traditional programming

Spam filter - ML Application technique
• What spam typically looks like?
• Spam filter is a Machine Learning program 1. Notice that some words or phrases (such as “4U,” “credit card,” “free,” and “amazing”) tend to
• Given examples of spam emails (e.g., flagged by users) and examples of regular (nonspam, also come up a lot in the subject line.
called “ham”) emails can learn to flag spam. • Perhaps a few other patterns in the sender’s name, the email’s body, and other parts of the email
• The examples that the system uses to learn are called the training set. could also be noticed

• Each training example is called a training instance (or sample). 2. Write a detection algorithm for each of the patterns noticed, and program would flag emails as
spam if a number of these patterns were detected.
• Task T is to flag spam for new emails
• 3. Test our program and repeat steps 1 and 2 until it was good enough
• Experience E is the training data
to launch.
• Performance measure P needs to be defined;
Problem?
• We can use the ratio of correctly classified emails.
• This particular performance measure is called accuracy, and it is often used in classification Program will likely become a long list of complex rules—hard to maintain
tasks.

Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron

Spam filter Application using traditional programming

technique Spam filter using ML Approach
• Write a spam filter using traditional programming techniques • a spam filter based on Machine Learning techniques automatically learns which words and
phrases are good predictors of spam
• by detecting unusually frequent patterns of words in the spam examples compared to the ham
examples
• The program is much shorter, easier to maintain, and most likely more accurate.

Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron

9
 31-08-2025

Spam filter using ML Approach Spam filter using ML Approach

• Write a spam filter using ML Approach • Automatically adapting to change
• If spammers keep working around our spam filter, we need to keep writing new rules forever
• Ex: What if spammers notice that all their emails containing “4U” are blocked?
• They might start writing “For U” instead.
• A spam filter using traditional programming techniques would need to be updated to flag “For U”
emails.
• In contrast, a spam filter based on Machine Learning techniques automatically notices that “For
U” has become unusually frequent in spam flagged by users, and it starts flagging them without
our intervention

Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron

Spam filter using ML Approach Spam filter using ML Approach

• Write a spam filter using ML Approach- Automatically adapting to change • Automatically adapting to change
• If spammers keep working around our spam filter, we need to keep writing new rules forever
• Ex: What if spammers notice that all their emails containing “4U” are blocked?
• They might start writing “For U” instead.
• A spam filter using traditional programming techniques would need to be updated to flag “For U”
emails.
• In contrast, a spam filter based on Machine Learning techniques automatically notices that “For
U” has become unusually frequent in spam flagged by users, and it starts flagging them without
our intervention

Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron

10
 31-08-2025

Other Applications using ML Approach Other Applications using ML Approach

• Another area where Machine Learning shines is for problems that either • Machine Learning can help humans learn
• are too complex for traditional approaches or • ML algorithms can be inspected to see what they have learned (although for some algorithms this
• have no known algorithm. can be tricky).
• Ex: Speech recognition • Ex: once a spam filter has been trained on enough spam, it can easily be inspected to reveal
• To write a program capable of distinguishing the words “one” and “two.” • the list of words and
• combinations of words that it believes are the best predictors of spam.
• We might notice that the word “two” starts with a high-pitch sound (“T”)
• Sometimes this will reveal unsuspected correlations or new trends, and thereby lead to a better
• so we could hardcode an algorithm that measures high-pitch sound intensity and use that to
distinguish ones and twos understanding of the problem.

• Problem? • Applying ML techniques to dig into large amounts of data can help discover patterns that were
not immediately apparent.
• But this technique will not scale to thousands of words spoken by millions of very different people
in noisy environments and in dozens of languages. • This is called data mining

• The best solution is to write an algorithm that learns by itself, given many example recordings for
each word.

Other Applications using ML Approach Why Use Machine Learning?- Summary

Machine Learning can help humans learn Machine Learning is great for:
• Problems for which existing solutions require a lot of fine-tuning or long lists of rules:
• one Machine Learning algorithm can often simplify code and perform better than the
traditional approach.
• Complex problems for which using a traditional approach yields no good solution:
• the best Machine Learning techniques can perhaps find a solution.
• Fluctuating environments: a Machine Learning system can adapt to new data.
• Getting insights about complex problems and large amounts of data.

11
 31-08-2025

EXAMPLES OF APPLICATIONS IN DIVERSE FIELDS EXAMPLES OF APPLICATIONS IN DIVERSE FIELDS

• Machine learning is a growing technology used to mine knowledge from data (popularly known as
data mining field).
• Wherever data exists, things can be learned from it.
• Whenever, there is excess of data, the mechanics of learning must be automatic.
• Machine learning technology is meant for automatic learning from voluminous datasets.
• Applications emerge not from machine learning experts, nor from the data itself, but from people
who work with the data and the problems from which it arises.

Applied Machine Learning - Gopal

If we download a copy of Wikipedia, has my computer really learned

something? Is it suddenly smarter? ML Future
• If we just download a copy of Wikipedia, our computer has a lot more data, • Hopefully soon there will be safe and efficient self-driving cars
• but it is not suddenly better at any task. Thus, downloading a copy of Wikipedia is not Machine • Notable progress has been made in medical applications;
Learning. • researchers demonstrated that deep learning models can detect skin cancer with near-
human accuracy ([Link]
• Another milestone was recently achieved by researchers
• DeepMind, who used deep learning to predict 3D protein structures, outperforming physics-
based approaches for the first time ([Link]

Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow Aurélien Géron Python Machine Learning- Machine Learning and Deep Learning with Python, scikit-learn, and TensorFlow 2
Sebastian Raschka, Vahid Mirjalili

12
 31-08-2025

FORMS OF LEARNING FORMS OF LEARNING

• In the broadest sense, any method that incorporates information from experience in the design of
a machine, employs learning.
• A learning method depends on the type of experience from which the machine will learn (with
which the machine will be trained).
• The type of available learning experience can have significant impact on success or failure of the
learning machine.
• The field of machine learning usually distinguishes four forms of learning:
1. supervised learning,
2. unsupervised learning,
3. reinforcement learning,
4. learning based on natural processes—evolution, swarming, and immune systems

FORMS OF LEARNING – Supervised Learning FORMS OF LEARNING – Supervised Learning

A supervised learning workflow
• the labeled training data is passed to a machine learning algorithm for fitting a predictive model that • The main goal in supervised learning is to learn a model from labeled training data that allows us
can make predictions on new, unlabeled data inputs: to make predictions about unseen or future data.
• Here, the term "supervised" refers to a set of training examples (data inputs) where the desired
output signals (labels) are already known.

13
 31-08-2025

FORMS OF LEARNING – Supervised Learning FORMS OF LEARNING – Supervised Learning

A supervised dataset
• A problem in which we are required to predict a value is known as a supervised problem.
• Ex1:
• if the problem is to predict house prices given historical house prices, with features like presence
of a hospital, school or supermarket, distance to nearest public transport, etc. is a supervised
problem.
• Ex2:
• Similarly, when we are provided with images of cats and dogs, and we know beforehand which
ones are cats and which ones are dogs, and if the task is to create a model which predicts
whether a provided image is of a cat or a dog, the problem is considered to be supervised.

Approaching (Almost) Any Machine Learning Problem – Abhishek Thakur Approaching (Almost) Any Machine Learning Problem – Abhishek Thakur

FORMS OF LEARNING – Supervised Learning FORMS OF LEARNING – Unsupervised Learning

An unsupervised dataset
• In figure, every row of the data is associated with a target or label.
• The columns are different features and rows represent different data points which are usually
called samples.
• The example shows ten samples with ten features and a target variable which can be either a
number or a category.
• If the target is categorical, the problem becomes a classification problem.
• If the target is a real number, the problem is defined as a regression problem.
• Supervised problems can be divided into two sub-classes:
• Classification: predicting a category, e.g. dog or cat.
• Regression: predicting a value, e.g. house prices.

Approaching (Almost) Any Machine Learning Problem – Abhishek Thakur Approaching (Almost) Any Machine Learning Problem – Abhishek Thakur

14
 31-08-2025

FORMS OF LEARNING – Unsupervised Learning FORMS OF LEARNING – Unsupervised Learning

• Unsupervised datasets do not have a target associated with them • When we know the number of clusters, we can use a clustering algorithm for unsupervised
• Ex1: problems.

• A financial firm deals with credit card transactions. • In figure, the data is assumed to have two classes,
• Dark colour represents fraud, and
• There is a lot of data that comes in every second.
• light colour represents genuine transactions.
• The only problem is that it is difficult to find humans who will mark each and every transaction
• These classes, however, are not known to us before the clustering approach.
either as a valid or genuine transaction or a fraud.
• After a clustering algorithm is applied, we should be able to distinguish between the two
• When we do not have any information about a transaction being fraud or genuine, the problem
becomes an unsupervised problem. assumed targets.

• To tackle these kinds of problems we have to think about how many clusters can data be divided • To make sense of unsupervised problems, we can also use numerous decomposition techniques
into. such as Principal Component Analysis

• Clustering is one of the approaches for problems but there are several other approaches • (PCA), t-distributed Stochastic Neighbour Embedding (t-SNE) etc.

• For a fraud detection problem, we can say that data can be divided into two classes (fraud or
genuine).
Approaching (Almost) Any Machine Learning Problem – Abhishek Thakur Approaching (Almost) Any Machine Learning Problem – Abhishek Thakur

FORMS OF LEARNING – Unsupervised Learning FORMS OF LEARNING – Reinforcement Learning

• Reinforcement learning is founded on the concept that if an action is followed by a satisfactory
state of affairs, or by an improved state of affairs (according to some properly defined way), then
the inclination to produce that action becomes stronger, i.e., reinforced.
• This idea can be extended to permit action choices to be dependent on state information, which
then brings in the aspect of feedback.
• A reinforcement learning system, therefore, is a system that via interaction with its environment
enhances its performance by obtaining feedback in the form of a scalar reward (or penalty)—a
reinforcement signal, that is indicative of the suitability of the response.
• The learning system is not instructed with regard to what action has to be taken.
• Instead, it is expected to find out which actions produce the maximum reward by trying them.
• The actions may influence not only the immediate reward but also the next situation, and
through that all subsequent rewards

Applied Machine Learning - Gopal

Approaching (Almost) Any Machine Learning Problem – Abhishek Thakur

15
 31-08-2025

Reinforcement Learning Reinforcement Learning

• Reinforcement learning is a feedback-based learning method, in which a learning agent gets a
reward for each right action and gets a penalty for each wrong action.
• The agent learns automatically with these feedbacks and improves its performance.
• In reinforcement learning, the agent interacts with the environment and explores it.
• The goal of an agent is to get the most reward points, and hence, it improves its performance.
• The robotic dog, which automatically learns the movement of his arms, is an example of
Reinforcement learning.

FORMS OF LEARNING - Learning Based on Natural Processes:

Reinforcement Learning Evolution, Swarming, and Immune Systems

• In reinforcement learning, the goal is to develop a system (agent) that improves its performance • Some learning approaches take inspiration from nature for the development of novel problem
based on interactions with the environment. solving techniques.
• Since the information about the current state of the environment typically also includes a so- • The thread that ties together learning based on evolution process, swarm intelligence, and
called reward signal, we can think of reinforcement learning as a field related to supervised immune systems is that all have been applied successfully to a variety of optimization problems.
learning.
• Optimization may not appear to be like a machine learning task, but optimization techniques are
• However, in reinforcement learning, this feedback is not the correct ground truth label or value, commonly used as part of machine learning algorithms.
but a measure of how well the action was measured by a reward function.
• Through its interaction with the environment, an agent can then use reinforcement learning to
learn a series of actions that maximizes this reward via an exploratory trial-and-error approach or
deliberative planning.

Applied Machine Learning - Gopal

16