Machine Learning Algorithms in Depth

ML Algorithms in Depth: Bayesian Inference and Deep Learning

Chp02: Markov Chain Monte Carlo (MCMC)

• Estimate Pi: Monte Carlo estimate of Pi
• Binomial Tree Model: Monte Carlo simulation of binomial stock price
• Random Walk: self-avoiding random walk
• Gibbs Sampling: Gibbs sampling of multivariate Gaussian distribution
• Metropolis-Hastings Sampling: Metropolis-Hastings sampling of multivariate Gaussian mixture
• Importance Sampling: importance sampling for finding expected value of a function

Chp03: Variational Inference (VI)

• Mean Field VI: image denoising in Ising model

Chp04: Software Implementation

• Subset Generation: a complete search algorithm
• Fractional Knapsack: a greedy algorithm
• Binary Search: a divide and conquer algorithm
• Binomial Coefficients: a dynamic programming algorithm

Chp05: Classification Algorithms

• Perceptron: perceptron algorithm
• SVM: support vector machine
• SGD-LR: stochastic gradient descent logistic regression
• Naive Bayes: Bernoulli Naive Bayes algorithm
• CART: decision tree classification algorithm

Chp06: Regression Algorithms

• KNN: K-Nearest Neighbors regression
• BLR: Bayesian linear regression
• HBR: Hierarchical Bayesian regression
• GPR: Gaussian Process regression

Chp07: Selected Supervised Learning Algorithms

• Page Rank: Google page rank algorithm
• HMM: EM algorithm for Hidden Markov Models
• Imbalanced Learning: Tomek Links, SMOTE
• Active Learning: LR
• Bayesian optimization: BO
• Ensemble Learning: Bagging, Boosting, Stacking

Chp08: Unsupervised Learning Algorithms

• DP-Means: Dirichlet Process (DP) K-Means
• EM-GMM: EM algorithm for Gaussian Mixture Models
• PCA: Principal Component Analysis
• t-SNE: t-SNE manifold learning

Chp09: Selected Unsupervised Learning Algorithms

• LDA: Variational Inference for Latent Dirichlet Allocation
• KDE: Kernel Density Estimator
• TPO: Tangent Portfolio Optimization
• ICE: Inverse Covariance Estimation
• SA: Simulated Annealing
• GA: Genetic Algorithm

Chp10: Fundamental Deep Learning Algorithms

• MLP: Multi-Layer Perceptron
• LeNet: LeNet for MNIST digit classification
• ResNet: ResNet50 image search on CalTech101 dataset
• LSTM: LSTM sentiment classification of IMDB movie dataset
• MINN: Mult-Input Neural Net model for sequence similarity of Quora question pairs dataset
• OPT: Neural Net Optimizers

Chp11: Advanced Deep Learning Algorithms

• LSTM-VAE: time-series anomaly detector
• MDN: mixture density network
• Transformer: for text classification
• GNN: graph neural network

Environment

To install required libraries, please run the following commands:

python3 -m venv ml-algo

source ml-algo/bin/activate    //in linux
.\ml-algo\Scripts\activate.bat //in CMD windows
.\ml-algo\Scripts\Activate.ps1 //in Powershell windows

pip install -r requirements.txt

Manning Early Access Preview (MEAP)

This book is now available in Manning Early Access Preview.
Link to book: https://www.manning.com/books/machine-learning-algorithms-in-depth

It will help you develop mathematical intuition for classic and modern ML algorithms, learn the fundamentals of Bayesian inference and deep learning, as well as data structures and algorithmic paradigms in ML!

Citation

You are welcome to cite the book as follows:

@book{MLAlgoInDepth,
  author = {Vadim Smolyakov},
  title = {Machine Learning Algorithms in Depth},
  year = {2023},
  isbn = {9781633439214},
  publisher = {Manning Publications}
}