Two Core Texts. One Serious Learning Track.
Building real competence in machine learning and deep learning through study, implementation, experiments, and written evidence.

MASTER ROAD is a long-term learning track focused on building practical and theoretical competence in machine learning and deep learning.

This is not a reading list.
Each stage must leave visible evidence: notes, code, solved exercises, experiments, mistakes, and retrospectives.

Status: Active
Primary Text: An Introduction to Statistical Learning with Applications in Python

I am currently building a strong foundation in statistical learning before moving deeper into modern deep learning.

Current focus

• Supervised learning
• Model evaluation
• Bias-variance tradeoff
• Overfitting and regularization
• Linear models
• Classification
• Resampling methods
• Tree-based methods
• Ensembles
• Core statistical learning concepts

Goal

Build strong machine learning foundations through:

• structured reading notes
• Python implementation
• solved exercises
• small reproducible experiments
• written summaries and retrospectives

Status: Next
Primary Text: Dive into Deep Learning

After establishing the machine learning foundation, I will move into deep learning through hands-on implementation and model training.

Planned focus

• Neural networks
• Backpropagation
• Optimization
• Multilayer perceptrons
• Convolutional neural networks
• Modern deep learning workflows
• Practical model training
• Reproducible experiments

Goal

Connect theory to implementation through notebooks, small projects, model training, and experiment records.

Every stage must leave behind:

• structured reading notes
• solved exercises
• concept summaries
• implementation code or notebooks
• experiment records and result comparisons
• mistakes, limitations, and lessons learned
• a written retrospective after completion

The goal is not to simply read famous books and feel productive.

The goal is to turn study into evidence:

• concepts I can explain
• code I can run
• experiments I can reproduce
• mistakes I can learn from
• systems I can eventually build

This roadmap exists to make learning visible, disciplined, and cumulative.

• Master Roadmap
• ISLP Study Notes
• Dive into Deep Learning Study Notes

The first priority is depth, not speed.

For each chapter or topic, the minimum output should be:

1. read and summarize the core ideas
2. implement the important concepts in Python
3. solve selected exercises
4. record mistakes and confusions
5. write a short retrospective
6. connect the topic to future projects

Build a serious ML/DL foundation that can support:

• research-oriented project work
• medical AI experiments
• backend-integrated AI services
• reproducible model development
• stronger technical writing
• long-term engineering competence