Chapters are released as and when they are ready. Please subscribe here

Table Of Contents (draft, may get revised)

  1. 1st principles of Applied ML 

    1. Programming a Logic vs Learning a Logic: Why learn from data?

    2. Linear thinking vs non-Linear Thinking: The mental model for learning from data

    3. Inductive Vs Deductive reasoning:  The intuition behind learning from data.

    4. Statistical data analysis and ML: Inductive reasoning

    5. Correlation vs Causation: what the models actually learn?

    6. Mathematical Optimisation and Information Theory: How models learn from data?

    7. Design of Experiments and Applied ML: Controlled experiments for learning from data

    8. Probability theory and Bayesian Rule: Deductive reasoning

    9. Normal distribution: The  Darling of distributions

    10. Hypes, Myths and AutoML

  2. Principles & Practice 

    1. Problem understanding principles & practice

      1. Problem Framing and defining success 

        1. Supervised

          1. Point predictions (single or multi-label classification or regression)

        2. Self-Supervised

          1. Structured predictions a.k.a seq2seq modelling

        3. Reinforcement Learning

        4. Zero-shot task transfer

          1. When training can be avoided

      2. Data collection strategies

      3. Dataset validation​

        1. The Representativeness of the dataset

        2. Testing for Sampling bias vs sampling errors

        3. Human bias as sampling bias

      4. The Relevance of the dataset

      5. Quiz​

      6. Coding Exercise

    2. Data understanding principles & practice 

      1. Exploratory data analysis (EDA)

      2. The Measure of central tendency, dispersion, and frequency

      3. Handling Missing values

      4. Handling Outliers and extreme values

      5. Diagnosing Data Leakage

      6. Correlation: Linearity vs Non-linearity

      7. Encoding, Normalisation, and standardization

      8. Transformation, Normality and near normality

      9. Quiz​

      10. Coding Exercise

    3. Data Viz principles & practice

    4. Feature engineering principles & practice​​​​

    5. Learning principles & practice

    6. Model testing and tuning principles & practice

    7. Deploying and serving (scoring) principles & practice

    8. Continuous learning and Human in the loop principles & practice

    9. Model interpretation principles & practice

    10. MLOps 

      1. Iterative model development

      2. Choosing the right framework

      3. Kubeflow for MLOps

      4. ...

    11. The Latest and Greatest

      1. Survey of AutoML Packages, AutoVimL, AutoNLP

      2. Self-supervision: Encoder-Decoder architecture,

      3. Attention Mechanism

      4. Transformers

      5. N-Shot learning (Zero-shot, 1-shot)

      6. ...

    12. ​Code Snippets, Notebooks, and datasets