Courses / Artificial Intelligence

Home / Artificial Intelligence

Artificial Intelligence

Artificial Intelligence

ARTIFICIAL INTELLIGENCE: Ever wonder how machines learn, make decisions, and even recognize your preferences? This is the captivating realm of Artificial Intelligence (AI), a rapidly evolving field shaping our world in profound ways.

Saksham's Artificial Intelligence course offers a beginner-friendly yet comprehensive

approach, equipping you with the foundational knowledge to:

  • Understand the core concepts: Demystify fundamental AI concepts like machine learning, deep learning, and neural networks, and explore their diverse applications.
  • Navigate the AI landscape: Gain insights into various AI techniques, algorithms, and tools used to develop intelligent systems.
  • Unleash your creativity: Explore how AI is transforming various industries, from healthcare and finance to transportation and entertainment, and spark your imagination for future possibilities.

Why Explore AI?

Here's why understanding AI is valuable in today's world:

  • Future-proof your career: Equip yourself with in-demand skills relevant across various industries, enhancing your career prospects.
  • Enhance critical thinking: Develop your ability to analyze information, solve problems, and think creatively, valuable skills beyond AI.
  • Become a responsible citizen: Gain insights into the ethical considerations surrounding AI and its impact on society.

Shaksham's AI course provides an engaging and interactive learning experience:

  • Expert-Led Video Lectures: Gain clarity with concise and engaging videos delivered by experienced instructors.
  • Interactive Exercises and Projects: Apply your knowledge through practical exercises and explore real-world AI applications through stimulating projects.
  • Supportive Learning Community: Connect and collaborate with fellow learners and instructors in a friendly and encouraging environment.

Unlock the fascinating world of AI with Saksham! By enrolling in this course, you'll gain the foundational knowledge and ignite your curiosity to explore the future of intelligent technologies and their impact on our lives.

Start your AI journey today with Saksham!


Course Overview

Python Basics
  • Introduction to Python
  • Basic Syntax
  • Data Types
  • Variables,
  • Operators
  • Input/output
  • Declaring variable
  • data types in programs o Your First Python Program
  • Flow of Control (Modules, Branching)
  • If, If- else, Nested if-else
  • Looping, For, While
  • Nested loops
  • Control Structure
  • Uses of Break & Continue
In Python, a string is a sequence of characters enclosed in single quotes ('') or double quotes (""). It is used to represent textual data.
  • Pass, Strings and Tuples
  • Accessing Strings
  • Basic Operations
  • Assigning Multiple Values at Once
  • Formatting Strings
  • String slices
Dictionaries
  • Introducing Dictionaries
  • Defining Dictionaries
  • Modifying Dictionaries
  • Deleting Items from Dictionaries
Python, a list is a versatile data structure that allows you to store and manipulate a collection of items.
  • Working with Lists
  • Introducing Lists o Defining Lists
  • Declare, assign and retrieve values from Lists
  • Accessing list
  • Operations in Lists
  • Adding Elements to Lists
  • Searching Lists
  • Deleting List Elements
  • Using List Operators
  • Mapping Lists
  • Joining Lists and Splitting Strings
  • Historical Note on String Methods
Python, a function is a block of reusable code that performs a specific task.
  • Function and Methods
  • Defining a function
  • Calling a function
  • Types of functions
  • Function Arguments
  • Anonymous functions
  • Global and local variables
  • Using Optional and Named Arguments
  • Using type, str, dir, and Other Built-In Functions
Tuple
  • Working with Tuples
  • Introducing Tuples
  • Accessing tuples
  • Operations
Advanced Python:
  • Object Oriented Python
  • OOPs concept o What's an Object?
  • Indenting Code o Native Data types
  • Declaring variables
  • Referencing Variables
  • Object References
  • Class and object
  • Attributes, Inheritance
  • Overloading & Overriding
  • Data hiding
  • Regular Expressions Using python
  • Object Oriented Linux Environment
Python, an exception is an event that occurs during the execution of a program that disrupts the normal flow of the program's instructions.
  • Operations Exception
  • Exception Handling
  • Except clause
  • Try finally clause
  • User Defined Exceptions
Python has a vast ecosystem of libraries and modules that extend the functionality of the language.
  • Python Libraries
  • Libraries and Functionality Programming
  • Debugging basics
Python provides powerful tools for data manipulation, analysis, and cleaning.
  • Working with Pandas
  • Data wrangling with Pandas
  • Working with NumPy , SciPy
  • Data cleaning with Python
Beautiful Soup is a Python library that is commonly used for web scraping and parsing HTML or XML documents.
  • Working with beautiful soup
  • Working with matplotlib, seaborn
  • Working with ggplot, plotly
- Descriptive statistics summarize and describe the main features of a dataset using measures like mean, median, mode, and standard deviation. - Inferential statistics make predictions or inferences about a population based on sample data using techniques like hypothesis testing and confidence intervals.
  • Introduction to Statistics- Descriptive Statistics
  • Summary Statistics - Central Tendency & Dispersion (Mean, Median, Mode, Quartiles, Percentiles, Range, Interquartile Range, Standard Deviation, Variance,Outliers and Coefficient of Variation)
  • Introduction to Inferential statistics
  • Sampling Techniques: Random Sampling and Stratified Sampling, Population and Sample, and Sampling Distributions
DATA SCIENCE
  • Overview of Data Analytics - Best Practices (this will also include understanding each student's
  • profile to align to their background & thought process)
  • Introduction to Data Analytics using some case studies
Python allows you to perform advanced numerical computations, data analysis, and visualization.
  • Working with NumPy, matplotlib, SciPy
data analysis that involves exploring and understanding the characteristics, patterns, and relationships present in a dataset.
  • Exploratory Data Analysis 1- Visualization and Exploring Data, Descriptive Statistical Measures,
  • Probability Distribution and Data
Data Analysis (EDA), let's explore two additional aspects: sampling and estimation, and statistical interfaces.
  • Exploratory Data Analysis 2: Sampling and Estimation, Statistical Interfaces
It is a key aspect of data analysis and machine learning.
  • Predictive modelling and analysis
Regression analysis is a statistical technique used to model and examine
  • Regression Analysis
Forecasting techniques are methods used to predict or estimate future values or trends based on historical data.
  • Forecasting Techniques
The uncertainty and variability in complex systems or processes.
  • Simulation and Risk Analysis
The practice of using quantitative techniques, data analysis, and mathematical modeling to support decision-making processes.
  • Decision Analytics,
  • Sensitivity Analysis
the creation of more sophisticated and informative visual representations of data.
  • Advanced Plots & Visualization
Strategy and analytics are two interconnected disciplines that play a crucial role in business decision-making and performance improvement.
  • Optimization Models - Linear, Non-Linear, Integer, Binary Programming + Cases
  • Strategy and Analytics
widely used technique in data analysis and machine learning for dimensionality reduction.
  • Overview of Factor Analysis,
  • PCA, Dimensionality Issues
Functional Data Analysis (FDA) is a statistical framework and methodology
  • Directional Data Analytics,
  • Functional Data Analysis
consider a case involving investment analytics for bonds.
  • Financial (TVM) Analytics
  • Investment (Bond) Analytics Case
  • Case studies: Making Right Business Decisions based on data
A multi-factor model is a financial model used to explain the returns of an investment or portfolio based on multiple factors or variables.
  • Multi Factor Model, Performance Measures
Portfolio analytics refers to the process of analyzing and evaluating the performance and characteristics of an investment portfolio.
  • Portfolio Analytics
  • Credit Risk Analytics & Fraud Analytics
  • Case studies: Making Right Business Decisions based on data
Web scraping refers to the automated extraction of data from websites.
  • Text analytics,
  • Social network analysis,
  • web scrapping,
Monte Carlo simulation is a computational technique
  • Dimensionality issues Ridge & lasso regression
  • Ridge & lasso regression
  • bias/variance trade off
  • Feature selection, Bagging and boosting
  • Simulation : Monte carlo
MACHINE LEARNING
  • What is machine learning?
  • Algorithm types of Machine learning
  • Supervised and Unsupervised Learning
  • Uses of Machine learning
  • Evaluating ML techniques
There are many algorithm under supervised and unsupervised machine learning.
  • REGRESSION:
  • Linear Regression Algorithm, Polynomial Regression, Decision Tree Regression, Random Forest Regression, Gradient Boosting Regression
  • CLASSIFICATION:
  • Logistic Regression, K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Decision Trees, Random Forest, Naive Bayes
  • CLUSTERING:
  • K-Means Clustering, Hierarchical Clustering, DBSCAN (Density-Based Spatial Clustering of Applications with Noise),
  • ASSOCIATION:
  • Apriori Algorithm
Classification and Regression Trees (CART) are a type of decision tree algorithm used for both classification and regression tasks.
  • Decision Trees
  • Classification and Regression Trees
The K-Nearest Neighbors (K-NN) algorithm is a simple and intuitive classification and regression algorithm.
  • Bayesian analysis and Naïve bayes classifier
  • Assigning probabilities and calculating results
  • Discriminant Analysis (Linear and Quadratic)
  • K-Nearest Neighbors Algorithm
Model Stacking are advanced machine learning techniques
  • Concept of Model Ensembling
  • Random forest, Gradient boosting Machines, Model Stacking
Apriori and FP-growth are two popular algorithms used for frequent itemset mining in data mining and association rule learning.
  • Association rules mining
  • Apriori and FP-growth algorithms
Support Vector Machines (SVMs) are powerful supervised machine learning models used for classification and regression tasks.
  • Basic classification principle of SVM
  • Support vector Machines
  • Linear and Non linear classification (Polynomial and Radial)
ARMA (Autoregressive Moving Average) and ARIMA (Autoregressive Integrated Moving Average) are popular models used for time series analysis and forecasting.
  • Moving average, Exponential Smoothing, Holt’s Trend Methods, Holt-Winters’ Methodsfor seasonality
  • Auto-correlation(ACF & PACF), Auto-regression, Auto-regressive Models, Moving Average Models
  • ARMA &ARIMA
It consists of an input layer, a single layer of artificial neurons (also called perceptrons or units), and an output layer.
  • Neural Network and its applications o o o
  • Single layer neural Network
  • Activation Functions: Sigmoid, Hyperbolic Tangent, ReLu
  • Overview of Back propagation of errors
Deep learning is that it is the branch of machine learning that is based on artificial neural network architecture.
  • Introduction to Deep Learning
  • Introduction to Convolutional Neural Network & Recurrent Neural Network
  • Introduction to Auto-encoders
Introduction to Deep Neural Network
  • Introduction Deep Learning and Neural Networks
  • Practical Application of Neural Networks
  • "Hello World" of Neural Network (Logistic Regression)
  • Cost Function
  • Activation
  • Gradient Descent for Logistic Regression
  • Python and NumPy primer for Deep Learning
If the derivative is positive, the function is increasing at that point.
  • Python and numpy refresher
  • Derivatives refresher
To understand a dataset, there are several key aspects you can explore.
  • Understand the dataset : Make Predictions (to be done in Jupyter notebook. Need Jupyter-notebook and libraries installed)
  • Implement your first Forward and Backward propagation
  • Implement activation function, gradient descent
  • Build Neural Network Model
  • Test and optimize the model
  • Make Predictions (to be done in Jupyter notebook. Need Jupyter-notebook and libraries installed)
It is a fundamental aspect of human development and plays a crucial role.
  • Sigmoid Model
  • Sigmoid Loss function
  • Introduction to Learning Algorithm
  • Deriving the Gradient Descent Update Rule
  • Sigmoid Evaluation
  • Assignment –Lab:
  • Plotting Sigmoid 2D
  • Plotting Sigmoid 3D
  • Contour Plot
  • Plotting Loss
  • Standardization
  • Test/Train split
Deep learning is a subfield of machine learning that focuses on training and using neural networks.
  • Going Deeper in Neural Network
  • Shallow Neural Network
  • Hidden Units and Hidden Layers
  • Explore a few more activation functions (tanh, relu)
  • forward and backward propagation with a hidden layer
  • Deep Learning notations and Neural Network Representations
Regularization and dropout are techniques used in deep learning to prevent overfitting and improve the generalization ability of neural networks.
  • Parameters vs Hyper Parameters
  • Regularization, Dropouts
  • implement Regularization, Dropouts. Improve performance of the learning algorithms
L2 normalization and calculate the Frobenius norm, you can use various programming languages and libraries like Python with NumPy.
  • L2 normalization, frobenius norm
  • implement L2 normalization, frobenius norm
Gradient checking is a technique used to verify the correctness of the gradients computed during the training of a machine learning model.
  • Vanishing Gradient and Exploding Gradient problems
  • Gradient Checking
data normalization are all techniques commonly used in machine learning to improve model performance and prevent overfitting.
  • early stopping, dropouts and other methods for data normalization
Mini-batch gradient descent is a variation of gradient descent optimization algorithm commonly used in training deep learning models.
  • Optimization Algorithms, ADAM
  • Mini batch gradient descent
  • implement ADAM in python and tensorflow.
Gradient descent algorithms that are commonly used in training deep learning models.
  • RMSProp, Momentum and other gradient descent algorithms
TensorFlow is an open-source deep learning framework developed by Google.
  • Introduction to Tensorflow, Tensorflow data structures, Keras Library
  • Assignment –Lab:
  • Fashion MNIST and digits MNIST exercises using Keras and tensorflow.
Recurrent Neural Networks (RNNs) are a class of artificial neural networks that excel at processing sequential data.
  • Sequential Modelling, Introduction, basic building blocks, applications
  • Introduction to RNNs
Gates are mechanisms introduced in RNN architectures to control the flow of information within the network.
  • Introduction to gates, GRU and LSTM
Gradient problems refer to the challenges that arise during the training of deep neural networks
  • LSTM and GRU Pt 2 - deep dive. How LSTM and GRU resolves exploding and vanishing
  • gradient problems
Sequence modeling in the context of music generation refers to the task of training a deep learning model
  • Sequence Modeling Lab and python for music generation using deep learning.
  • Sequence Modeling - Word representation, word embedding matrix, learning words and embeddings, sentiment
  • classification
Reinforcement learning is an area of Machine Learning. It is the science of decision making.
  • Introduction to reinforcement learning as an approximate dynamic programming problem
  • Overview of reinforcement learning
Its potential for solving complex problems and learning optimal decision-making policies.
  • Success of reinforcement learning
  • The agent environment framework
MDPs are widely used in reinforcement learning and sequential decision-making problems.
  • Bandit problems and online learning
  • Markov decision processes
  • Solution methods
Returns and Value functions are essential concepts that help evaluate and estimate the long-term expected rewards associated with different states and actions.
  • Returns
  • Value functions
Dynamic programming is a technique used to solve optimization problems.
  • Dynamic programming
temporal differences between predictions and actual rewards.
  • Solution methods for temporal difference learning
Eligibility traces are a concept used in reinforcement learning to assign credit or importance to past state-action pairs in the learning process.
  • Eligibility traces
Value function approximation is a technique used in reinforcement learning
  • Value function approximation Models and planning (table lookup case),
Simulation-based methods like Q-learning are a class of reinforcement learning algorithms
  • Simulation based methods like Q-learning
Artificial Intelligence is a method of making a computer, a computer-controlled robot, or a software think intelligently like the human mind.
  • Introduction to AI
  • Why AI Now?
  • Revolution of AI
The evolution of AI (Artificial Intelligence) is a fascinating and rapidly progressing field
  • Philosophies of CS & AI,
  • AI Evolution: Turing's Work,
  • Turing Machine & Test
"Being Human in the Age of AI" is a thought-provoking topic that explores the implications
  • Revolution & Current Trends in AI,
  • Being Human in the Age of AI
AI have led to the development of various applications across different domains.
  • Introduction of Applications in various Domains (Scientific including Health Sciences, Engineering, Financial Services and other industries),
AI opponents in games and strategic decision-making scenarios.
  • Knowledge Representation, Problem Solving
  • Search Methodologies, Classical Search Methodologies,
  • Beyond Classical Search, Parallel Search, Search Engines
The structure of the game tree depends on the specific game being modeled.
  • Game Theory, various problems
  • Game Trees
Combinatorial optimization problems involve finding the best solution from a finite set of possibilities for a given optimization criterion.
  • Intelligent Agents, Uninformed Search,
  • Constraint Satisfaction Search,
  • Combinatorial Optimization Problems,
Resolution is a powerful reasoning technique used in automated theorem proving and logic programming.
  • Logic Concepts & Logic Programming
  • Inference and Resolution for Problem Solving
There are several libraries and programming functionalities available that provide additional functionality and capabilities to programming languages.
  • Rules and Expert Systems, ES Shells, Heuristic & Meta-heuristics
  • Python Libraries
  • Libraries and Functionality Programming
  • Debugging basics
The use of computer simulations, robotics, and other artificial means.
  • Artificial Life, Learning through,
  • Emergent Behavior,
  • Rules and Expert Systems,
Propositional Logic and Predicate Logic are two fundamental branches of formal logic .
  • Knowledge Representation and Automated,
  • Propositional and Predicate Logic,
  • Inference and Resolution for Problem Solving,
Monte Carlo planning employs random sampling or simulation techniques to estimate the expected outcomes of different actions or policies.
  • Advanced Problem Solving Paradigm: Planning
  • Planning Methods
Making decisions, whether simple or complex, involves a cognitive process of evaluating options, considering potential outcomes, and selecting the best course of action.
  • Advanced Planning Application: Case Study
  • Making Simple & Complex Decisions
Combinatorial optimization problems refer to a class of problems where the goal is to find the best possible solution from a finite set of possible solutions.
  • Combinatorial Optimization Problems,
  • Constraint Satisfaction Problems, Game Playing
  • Uncertainty Measure, Probabilistic Reasoning, PR over Time
  • Fuzzy Logic
Projects
  • Auto Spelling Correction
  • Anti Indian Statement Analysis
  • Date Tagging for DNN based Machine Translation
  • Custom NER for Indian Names with Spacy
  • Name Entity Reorganisation using Deep-Pavlov and BERT