Explore

Hema Kesana Masters Course

Managed by:

Ramakrushna⁠

⁠

AI 391L. Machine Learning

Focuses on foundational and theoretical aspects of machine learning, with a wide range of inductive classification techniques.

AI 391M. Case Studies in Machine Learning

Applies ML methods to practical, real-world data problems using case studies.

⁠

AI 391L—Machine Learning (Theory & Foundations)

1. Inductive Classification Techniques

Version space learning Concept learning using version spaces and hypotheses.

Decision Trees ID3, C4.5, CART, and pruning methods.

Rule-Based Systems Rule induction, RIPPER, CN2.

Neural Networks Perceptron, MLP, backpropagation, activation functions.

Bayesian Learning Naive Bayes, Bayesian networks, and probabilistic inference.

Instance-Based Learning k-NN, distance metrics, and weighting functions.

2. Computational Learning Theory

PAC learning

VC dimension

Bias-variance tradeoff

Generalization bounds

3. Explanation-Based Learning

Deriving generalizations from a single example using prior knowledge.

Combining symbolic reasoning with inductive learning.

4. Knowledge Refinement & Transfer

Incremental learning

Transfer learning basics

Concept drift and model updating

5. Learning Modalities

Supervised vs unsupervised

Online vs batch

Active learning

6. LISP and AI Fundamentals (Prerequisite Knowledge)

Programming basics in LISP

AI techniques like search, reasoning, and planning Prerequisite Knowledge Familiarity with:

Artificial Intelligence (CS 381K or equivalent)

LISP programming

AI 391M—Case Studies in Machine Learning (Application-Focused)

1. Review of Core ML Algorithms

Logistic regression, SVM, decision trees, random forests

Gradient boosting methods (e.g., XGBoost)

Neural networks, CNNs (for images), RNNs (for sequences)

2. ML Workflow with Real Data

Data preprocessing & cleaning

Feature engineering & selection

Model selection & hyperparameter tuning

Cross-validation, performance evaluation

3. Evaluation Techniques

Confusion matrix, ROC, precision/recall, F1 score

AUC, log-loss, calibration curves

Error analysis

4. Case Studies by Domain Examples (could vary by instructor):

Healthcare: Predicting disease risk

Finance: Fraud detection or credit scoring

Retail: Recommendation systems

Social Media: Sentiment analysis

Image/Video: Object detection or classification

5. Tools and Libraries

Python (NumPy, pandas, scikit-learn)

TensorFlow/PyTorch for deep learning

Jupyter notebooks for analysis

6. Project Work

Students analyze a dataset, apply ML, and present findings

Could include Kaggle-style competitions or open datasets

Skills Required

Skills Required

AI 391L

AI 391M

Mathematical rigor

Practical implementation

Theoretical grounding

Real-world application

Algorithm development

Model evaluation & deployment

Learning theory

Problem-solving using ML

There are no rows in this table

⁠

Book Recommendations

Book Recommendations

Topic

Recommended Book

Version Space, Decision Trees, Rule-Based Learning

Machine Learning by Tom M. Mitchell

Bayesian Learning

Pattern Recognition and Machine Learning by Christopher Bishop

Neural Networks

Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville

Instance-Based Learning (e.g., k-NN)

Machine Learning by Tom M. Mitchell

Computational Learning Theory

Understanding Machine Learning by Shai Shalev-Shwartz and Shai Ben-David

Explanation-Based Learning

Machine Learning by Tom M. Mitchell

Knowledge Refinement / Transfer Learning

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow by Géron

LISP Programming & AI Fundamentals

ANSI Common LISP by Paul Graham

AI Concepts & Search Methods

Artificial Intelligence: A Modern Approach by Stuart Russell & Peter Norvig

Core ML Algorithms & Evaluation

Introduction to Statistical Learning by Gareth James et al.

ML Workflow & Feature Engineering

Feature Engineering for Machine Learning by Alice Zheng

Deep Learning Application

Neural Networks and Deep Learning by Michael Nielsen

Real-World ML Projects & Pipelines

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow by Géron

Domain-Specific Case Studies

Applied Predictive Modeling by Max Kuhn

There are no rows in this table

⁠

Plan Structure (6-Months)

Phases

Phases

Phase

Duration

Focus

Phase 1

Month 1–2 (Weeks 1–8)

Core ML Foundations

Phase 2

Month 3–4 (Weeks 9–16)

Intermediate Models & Learning Theory

Phase 3

Month 5–6 (Weeks 17–24)

Case Studies, Model Deployment, Capstone

There are no rows in this table

⁠

Phase 1: Foundations (Weeks 1–8)

Phase 1: Foundations (Weeks 1–8)

Week

Topics

Resources

What is ML? Types, Workflow

Tom Mitchell Ch.1, Google ML Crash Course

Math Refresher: Linear Algebra, Probability

Khan Academy, 3Blue1Brown

Decision Trees, Rule-Based Learning

Mitchell Ch.3, Géron Ch.6

k-NN and Instance-Based Learning

Mitchell Ch.8, Géron Ch.3

Model Evaluation: Accuracy, F1, ROC

Géron Ch.3, sklearn

Feature Engineering & Data Cleaning

Feature Engineering by Alice Zheng

Intro to Neural Nets: Perceptron, Backprop

Nielsen, Géron Ch.10

Mini Project #1: Classification

Titanic Dataset / Breast Cancer (sklearn)

There are no rows in this table

⁠

Phase 2: Intermediate & Theory (Weeks 9–16)

Phase 2: Intermediate & Theory (Weeks 9–16)

Week

Topics

Resources

Deep Learning Basics: MLP, CNN intro

Goodfellow Ch.6–7, Fast.ai

Bayesian Learning: Naive Bayes, MAP

Bishop Ch.2, Géron Ch.4

Ensemble Methods: Bagging, Boosting

Géron Ch.7, XGBoost tutorials

Transfer Learning & Knowledge Refinement

Géron Ch.11, Hugging Face Docs

PAC Learning, VC Dimension

Understanding ML by Shai Shalev-Shwartz

Model Selection & Overfitting Theory

Géron Ch.5, Bias-Variance Intuition

Explanation-Based Learning

Mitchell's EBL chapters

Mini Project #2: NLP or Tabular Prediction

Use scikit-learn, or Hugging Face (NLP)

There are no rows in this table

⁠

Phase 3: Case Studies & Capstone (Weeks 17–24)

Phase 3: Case Studies & Capstone (Weeks 17–24)

Week

Topics

Tools

Case Study 1: Healthcare (e.g., Diabetes)

UCI dataset + sklearn

Case Study 2: Finance / Credit Risk

Kaggle dataset + XGBoost

Explainability: SHAP, LIME

SHAP library tutorials

ML Pipelines: Preprocessing to Deployment

scikit-learn pipeline + Streamlit

Capstone Project Planning (your own idea)

Topic proposal + dataset selection

Capstone Development – Model & Analysis

Code, evaluate, debug

Capstone Finalization – UI + Interpretation

Build simple demo w/ Streamlit or Gradio

Capstone Presentation

Publish to GitHub + optional blog post

There are no rows in this table

⁠

Want to print your doc?
This is not the way.

Try clicking the ··· in the right corner or using a keyboard shortcut (

CtrlP

) instead.