Skip to content

Blog Ideas

I would like to do some blog posts for my academic blog. There are tons of things I would like to do but I don't really have time to do it all. So I would like to try a concept: combine scientific exploration with some programming concept exploration. For example, instead of doing a segment on Object-Oriented Programming and then a separate segment on Gaussian Process regression (GPR), I would combine the two.

Programming

Packages

  • GPyTorch - DKL,
  • Pyro
  • GPFlow
  • geopandas
  • xarray

Algorithms

  • Optimized Kernel Ridge Regression (OKRR)
  • Jax
  • Optimized Kernel Entropy Components Analysis (OKECA)
  • Jax
  • Rotation-Based Iterative Gaussianization (RBIG)
  • Gaussian Process Regression
  • Variational Gaussian Process Regression
  • Variational Inference
  • Jax
  • Bayesian Neural Networks
  • Deep Kernel Learning
  • GPyTorch - Gaussian Processes
  • PyTorch - Neural Networks
  • Data - Ocean Water Quality
  • Gaussian Processes
  • Exact
  • Variational
  • Sparse
  • Sparse Variational
  • Deep

Packages

  • Edward2 - Bayesian Layers (the future)

Unsorted

  • Bayesian Formulas + Plotly
  • OOP + GPR from Scratch
  • Kernel Functions (K, GPR, KRR) + Derivatives + AutoGrad
  • VI + PyTorch
  • RBIG + sklearn API
  • IT Measures + ESDC + RBIG
  • AD + ESDC + Feature Selection
  • AD + ESDC + Feature Selection (Pt II)
  • HyperLabelMe + TPOT + AutoSklearn
  • Flask + Xarray
  • Luigi + SLURM + Experiment
  • Uncertainty + GPs
  • Abstract Classes + Kernel Functions

Work Deep Dive

These notebooks will be directly related to my thesis and things that I am investigating actively. They should all include some results so that I can show off some of the actual applications.

  • Output Normalized Methods
  • I - Kernel Eigenmap Methods
  • II - Kernel Eigenmap Projection Methods
  • III - Manifold Alignment
  • IV - Nearest Neighbours (Annoy, KDE Trees)
  • V - Eigenvalue Decomposition Scaling (rSVD, Multigrid, Random Projections)
  • VI - Out of Sampling (Nystrom, LLL, Var. Nystrom)
  • Kernel Methods
  • I - Kernel Functions
  • II - Learning with Kernel Functions (Overview of Literature)
  • III - Gradients and Sensitivity Analysis
  • Gaussian Processes and Uncertainty
  • I - GPs
  • II - Sparse GPs
  • III - Uncertain GPs (Literature, NIGP, My Work)
  • IV - Variational Methods
  • Deep Density Destructors
  • I - Density Estimation
  • II - RBIG
  • III - GDN

Concept Notebook

These are notebooks that I decided to investigate because either I sucked at in the beginning or it was something I needed in order to advance to the next level of whatever I was doing related to my thesis.

  • Bayesian Methods
  • Variational Inference
  • Information Theory
  • Anomaly Detection

Explorers Book

  • Normalizing Flows
  • Automatic Machine Learning
  • Neural ODEs

Lab Notebook

  • Earth Science Data Cube + Xarray
  • Naive AD Detection
  • Dask
  • Remote Computing

xarray

  • Shape Files
  • Region Masks
  • Large Scale ML - PCA, LR, KMeans, XGBoost
  • Dask

Jax

  • Kernel methods
  • kernels
  • regression - krr, rff
  • bayesian regression - gp, sgps
  • classification - svm
  • dependence estimation - hsic
  • dimension reduction - okeca)
  • Gaussianization flows