Embarking on the
Machine Learning
Journey: A Beginner's Guide
Supriya Khadka
Conversational AI Engineer, [Link]
Agenda
1 2 3
What is Machine What can I do How do I get
Learning? with it? started?
What were my four years like?
Course Work
Year I
What were my four years like?
Course Work Web Development
Year I Year II
What were my four years like?
Course Work Web Development Blockchain
Year I Year II Year III
What were my four years like?
Course Work Web Development Blockchain Machine Learning
Year I Year II Year III Year IV
What is Machine Learning?
Machine Learning
Allowing machines to learn from data and algorithms
Enabling it to make decisions without being explicitly
programmed
Traditional Programming vs Machine Learning
Traditional Programming
Data
Computation Output
Program
Traditional Programming vs Machine Learning
Machine Learning
New Data
Data Program
Training Output
Desired Output (Model)
Machine Learning Example
Program
Training Apple
(Model)
Labelled Fruit Collection
Machine Learning Example
Program
Training ??
(Model)
Labelled Fruit Collection
AI vs ML vs DL
Rule-based System,
Expert System
Deep Learning
(Neural Nets) ML AI
Machine Learning vs Deep Learning
Output
ML
Raw Data Feature Extraction
Multiclass Classification
Output
DL
Raw Data Feature Extraction +
Classification
Semi-Supervised
Supervised Unsupervised Reinforcement
Labelled Data Unlabelled Data Make decisions based
on the environment
Apple
Or
an
ge
Pear
Types of Machine Learning
Machine Learning Domains
Natural Language Processing
Text Classification
Sentiment Analysis
Language Modelling
Conversational AI
Machine Translation/Transliteration
Named Entity Recognition
Computer Vision
Object Detection
Image Classification
Semantic Segmentation
Object Tracking
Facial Recognition
Medical Imaging
Signal Processing
Speech Recognition
Text-to-Speech Synthesis
Image Processing
Time Series Analysis
Biomedical Signal Processing
Sensor Data Processing
Robotics/ Simultaneous Localization
and Mapping (SLAM)
Visual SLAM
Lidar SLAM
Deep Reinforcement Learning
for Navigation
Localization and Mapping in
GPS-denied Environments
Data Science
Customer Segmentation
Fraud Detection
Predictive Maintainence
Recommender Systems
Many of the examples
mentioned previously
Let's Play A Game!
Machine Learning: Necessity or Overkill?
Sorting a list of Numbers?
Sorting a list of Numbers?
Overkill
Speech Recognition
Speech Recognition
Necessity
Spam Email Detection
Spam Email Detection
Necessity*
Spam Email Detection
Necessity*
*unless you have predefined
thresholds
Insurance Claim Verification
Insurance Claim Verification
Underkill(?)*
Insurance Claim Verification
Expert Systems or People
*unless you have fixed predefined
criteria
Cervical Cancer Detection
Cervical Cancer Detection
Necessity*
*consider size of datasets and only
use as Decision Support System
How to Get Started?
Programming Language Mathematics Fundamentals
1 2
Python Programming Language Linear Algebra - Gilbert Strang
Basics of Python Programming Vectors, Matrices
Variables and Data Types Calculus
Control Flow Differentiation, Integration
Functions and OOP Probability and Statistics
Probability Theory
Statistical Distributions
Hypothesis Testing
Documentation
Regression Analysis
Programiz, FreeCodeCamp
Coursera - Python for Everybody
Engineering Course Content
Literally Everywhere on the Internet
Andrew NG - Coursera
Gilbert Strang (MIT) - Linear
Algebra
Machine Learning Concepts Machine Learning Algorithms
3 4
Types of Machine Learning Linear Regression, Logistic Regression,
Model Training Support Vector Machines, Decision
Gradient Descent, Regularization Tree, Random Forest, Clustering, PCA
Model Evaluation and Validation Neural Network Concepts
Active Learning CNNs, RNNs, LSTMs, Transformers
Transfer Learning Intuition Behind each algorithm
Strengths and Weaknesses
Andrew NG - Coursera
MachineLearningMastery Youtube Videos
"Pattern Recognition and Machine Implement after studying!
Learning" by Christopher Bishop* [Link]
"Deep Learning" by Ian Goodfellow,
*step with caution Yoshua Bengio, and Aaron Courvill
5 Machine Learning Libraries and Frameworks
[Link], Official Documentation and
Youtube Tutorials, Hugging Face
"Hands-On Machine Learning with Scikit-
Scikit-Learn
Learn, Keras, and TensorFlow" by Aurélien
Tensorflow, Keras
Géron
Pytorch "Machine Learning with PyTorch and Scikit-
Learn" Sebastian Raschka
Open Source Projects/Kaggle
5 Machine Learning Libraries and Frameworks
[Link], Official Documentation and
Youtube Tutorials, Hugging Face
"Hands-On Machine Learning with Scikit-
Scikit-Learn
Learn, Keras, and TensorFlow" by Aurélien
Tensorflow, Keras
Géron
Pytorch "Machine Learning with PyTorch and Scikit-
Learn" Sebastian Raschka
Open Source Projects/Kaggle
Your cue to start a project!
Data Preprocessing Model Evaluation
6 7
Feature Engineering Cross Validation
Handling Missing Values Performance Metrics
Scaling Features Accuracy, Precision, Recall, F1
Handling Categorical Values Score, AUC-ROC
Extracting Meaningful Information Mean Squared Error, Mean
Absolute Error
Overfitting/Underfitting
BIas-Variance Trade-Off
"Feature Engineering for Machine
Learning" on Coursera
See a lot of other projects to Books and Courses mentioned before
understand how it's done: Kaggle, have all of this information
Medium Blogs Continue the project: You will get
Hands-on Experience some ideas while implementing
8 Practice and Projects
Work on real-world Machine Learning Competitions on
datasets [Link]
Implement different Papers with Code
algorithms Read Research Papers (Harvard CS197)
Evaluate model Open Source Projects
performance Make your own Project
Start a Project!
How to start your ML Project?
Define the Collect Explore Data Pre- Split the
Problem Data the Data processing Data
First Strategy
Deploy Finetune Evaluate Train the Choose a
the model the model the model model Model
You got this!
You got this!
Start Today!