Foundations of Machine Learning Algorithms: Pen-Paper Calculations
" Foundations of Machine Learning (Algorithms): Pen-paper calculations " course is a non-coding course, which is a MUST for ALL persons desirous of learning Machine Learning from mathematical and algorithmic point of view. We will focus more on the theoretical aspects of the algorithms, parameters and hand-calculations will be done on dummy data step-by-step. In some cases, to automate the calclations, we will be using MS Excel.
The Following algorithms are tentatively planned to be discussed and detailed tutorials/examples will be worked out in the class.
- General Introduction:
- Parametric and Non-parametric Machine Learning Algorithms
- The Supervised, Unsupervised and semi-supervised Learning
- The Bias-Variance Trade-off
- Overfitting and Underfitting
Essential Mathematics for Machine Learning - Part – A: Basic Probability
- Basic Definitions
- Even & odds of an event
- Bayes Theorem & applications
- Probability Distribution Functions
- Mean, Mode, Median
- Standard Deviation, Variance
- Correlation and Correlation-coefficient
- Standard Statistical Distributions
Essential Mathematics for Machine Learning - Part – C: Linear Algebra
- Matrix Multiplication
- Operations and Properties
- Identity Matrix and Diagonal Matrices
- Transpose, Inverse, Trace, Norms and Determinant of Matrices
- Symmetric & Orthogonal Matrices
- Linear Independence and Rank
- Eigenvalues and Eigenvectors of Symmetric Matrices
- Matrix Calculus
- Gradients and Hessians of Quadratic and Linear Functions
- Least Squares
- Gradients of the Determinant
- Eigenvalues as Optimization
- Linear Algorithms:
- Gradient Descent.
- Linear Regression.
- Logistic Regression.
- Linear Discriminant Analysis.
- Non-Linear Algorithms:
- Classification and Regression Trees.
- Naive Bayes.
- K-Nearest Neighbors.
- Learning Vector Quantization.
- Support Vector Machines.
- Ensemble Methods:
- Bagged Decision Trees and Random Forest.
- Boosting and AdaBoost.
- Clear algorithm explainations that help you to understand the principles that underlie each technique.
- The step-by-step algorithm workout on black-board to show you exactly how each model learns.
- Real worked examples so that you can see exactly the numbers in and the numbers out, there’s nowhere for the details to hide.
Training Schedule:Saturday-Sunday Batch
|6 Days||17, 24 Sept.; 1, 7, 8, 14 October 2017|
|Saturday Time:||9:30 AM - 1:00 PM|
|Sunday Time:||8:00 AM – 11:00 AM|
Page Last Updated: Saturday 02-Sep-2017 02:48:42 IST