This is an ever changing to-do list of topics I would like to learn and notes to create.

🚧 is where my focus currently is.

Current

Dimensionality reduction 🚧

Classic linear methods

• PCA 🚧
• probabilistic PCA
• Bayesian PCA
• Factor Analysis (FA) 🚧
  • Some examples from Neuroscience?
• Gaussian Process Factor Analysis (GPFA) 🚧
• Canonical correlation analysis (CCA)
• jPCA
• seqPCA

! [[ for_exceptional_broski.png ]]

Regression?

• Partial least squares (PLS)
• Fisher’s linear discriminant (FLD)
• Linear discriminant analysis (LDA)
• Principal component regression (PCR)
• Independent component analysis (ICA)
• GLM methods for dimensionality reduction?

Targeted approaches

Methods that seek subspaces that are relate to behavior or task variables (i.e., measured experimental variables)

• Demixed PCA
• Targeted dimensionality reduction (TDR)
• Model-based targeted dimensionality reduction (mTDR)
• Preferential subspace identification (PSID)
• Subspace identification for linear systems (SID)

Latent dynamics

• LFADS
• VAEs

Nonlinear methods

• Kernel PCA
• Isomap
• T-SNE
• UMAP

• Multi-dimensional scaling

• GLM methods for dimensionality reduction

Future

• Fix, expand, and post predictive coding notes
• Topological data analysis
  • persistent homology
  • mapper algorithm
• Causal inference
• Make notes on different bursting neuron models
  • Include some simulations in the notes
  • Include a note on my own paper
• Notes on point-process models