Schedule

Week 1

Jan 3 - 4

Saturday 6.00 p.m.–8.00 p.m.

Introduction & Tools Setup

Program overview, expectations, assessments
Introduction to AI / ML / Deep Learning — definitions, history, use‑cases
Environment setup: Python, Jupyter, key libraries (NumPy, pandas, scikit‑learn, PyTorch/TensorFlow)

Sunday 11.00 a.m.–1.00 p.m.

Python for Data Science / Linear Algebra Refresher

• Numpy, pandas, data loading, basic manipulations

• Linear algebra recap (vectors, matrices, operations)

• Simple exercises (vector ops, matrix multiplications)

Assignment: small data handling exercises; reading material on ML pipeline

Week 2

Jan 10 - 11

Saturday 6.00 p.m.–8.00 p.m.

Supervised Learning — Regression & Classification • Linear Regression, logistic regression • Loss functions, training, gradient descent

Sunday 11.00 a.m.–1.00 p.m.

Model Evaluation & Overfitting / Regularization • Train/test split, cross‑validation • Metrics: RMSE, MAE, accuracy, precision/recall, ROC AUC • Bias‑variance tradeoff, regularization (L1, L2)

Assignment: build regression & classification models on sample datasets (e.g. housing, credit data)

Week 3

Jan 17 - 18

Saturday 6.00 p.m.–8.00 p.m.

Advanced Models: Tree-based & Ensemble Methods • Decision Trees, Random Forests, Gradient Boosting (XGBoost / LightGBM / CatBoost)

Sunday 11.00 a.m.–1.00 p.m.

Feature Engineering & Model Tuning

• Feature selection, encoding, scaling

• Hyperparameter tuning (grid search, random search)

• Introduction to pipelines

Assignment: participate in a small Kaggle-style mini problem with tree ensembles

Week 4

Jan 24 - 25

Saturday 6.00 p.m.–8.00 p.m.

Neural Networks & Deep Learning Foundations • Perceptron, MLPs, activation functions • Backpropagation, optimization methods (SGD, Adam)

Sunday 11.00 a.m.–1.00 p.m.

Deep Learning Frameworks & Hands-on • Build simple feedforward nets in PyTorch (or TensorFlow) • Training / validation / overfitting in DL • Early stopping, dropout, batch norm

Assignment: build a neural net for a classification / regression problem; compare with classical models

Week 5

Jan 31 - Feb 1

Saturday 6.00 p.m.–8.00 p.m.

Computer Vision / Convolutional Neural Networks (CNNs) • Convolution, pooling, architectures (LeNet, AlexNet, VGG) • Transfer learning, pretrained models

Sunday 11.00 a.m.–1.00 p.m.

Image Processing Techniques & Advanced CV • Data augmentation, object detection basics (e.g. YOLO, SSD) • Segmentation basics (UNet) introduction

Assignment: fine‑tune a pretrained CNN to a custom image dataset (e.g. classify among a few classes)

Week 6

Feb 7 - 8

Saturday 6.00 p.m.–8.00 p.m.

Natural Language Processing (NLP) & Language Models • Text processing (tokenization, embeddings) • Word2Vec, GloVe, contextual embeddings

Sunday 11.00 a.m.–1.00 p.m.

Transformer Models & Large Language Models (LLMs) • Transformer architecture, attention mechanism • BERT / GPT / encoder‑decoder models • Prompt engineering basics

Assignment: fine‑tune a small transformer (or use HuggingFace) for sentiment classification / Q&A task

Week 7

Feb 14 - 15

Saturday 6.00 p.m.–8.00 p.m.

Generative Models & Advanced Topics • Variational Autoencoders (VAEs), GANs • Diffusion models (introduction)

Sunday 11.00 a.m.–1.00 p.m.

Reinforcement Learning & Ethics / Safety in AI • RL basics (MDP, Q‑learning, policy gradients) • AI ethics, fairness, bias, explainability, privacy

Assignment: build a simple generative model (e.g. VAE) or explore reinforcement in a toy environment

Week 8

Feb 21 - 22

Saturday 6.00 p.m.–8.00 p.m.

Capstone Project Kickoff / Design Workshop • Introduction of possible capstone themes • Team formation, project proposals, data sources, architecture design

Sunday 11.00 a.m.–1.00 p.m.

Capstone Project Implementation & Mentoring • Initial implementation, setup, milestone plan • Guidance, troubleshooting

After Week 8, students continue working (weekly check-ins or office hours) and present at the end on Week 9.

Page updated

Report abuse