Incrementally Update a SONG Model with New Data

Adds new data to an existing SONG model. This is the key differentiator of SONG over t-SNE and UMAP: the existing embedding is preserved while new coding vectors and edges are grown to accommodate the new data.

Usage

# S3 method for class 'song_model'
update(object, X_new, epochs = 50L, alpha = NULL, verbose = TRUE, ...)

Arguments

object

A "song_model" object to update.

X_new

Numeric matrix of new data (n_new x D). Must have the same number of columns as the original training data.

epochs

Integer. Number of additional training epochs (default: 50).

alpha

Numeric or NULL. Initial learning rate for the update. If NULL, uses the original learning rate (default: NULL).

verbose

Logical. Whether to print progress (default: TRUE).

...

Ignored.

Value

An updated "song_model" object with grown codebook incorporating the new data.

References

Senanayake, D. A., Wang, W., Naik, S. H., & Halgamuge, S. (2021). Self-Organizing Nebulous Growths for Robust and Incremental Data Visualization. IEEE Transactions on Neural Networks and Learning Systems, 32(10), 4588–4602. doi:10.1109/TNNLS.2020.3023941

See also

song, predict.song_model

Examples

model <- song(as.matrix(iris[1:100, 1:4]), epochs = 5L, seed = 42)
#> Epoch 1/5 | CVs: 3 | QE: 1.2359 | so_lr: 1.0000 | lr: 1.0000
#> Epoch 2/5 | CVs: 5 | QE: 0.6649 | so_lr: 0.8187 | lr: 0.8000
#> Epoch 3/5 | CVs: 9 | QE: 0.4649 | so_lr: 0.4493 | lr: 0.6000
#> Epoch 4/5 | CVs: 16 | QE: 0.3477 | so_lr: 0.1653 | lr: 0.4000
#> Epoch 5/5 | CVs: 21 | QE: 0.3056 | so_lr: 0.0408 | lr: 0.2000
#> Running UMAP dispersion step...
model <- update(model, as.matrix(iris[101:150, 1:4]), epochs = 5L)
#> Update Epoch 1/5 | CVs: 21 | QE: 0.5691 | so_lr: 1.0000
#> Update Epoch 2/5 | CVs: 32 | QE: 0.4294 | so_lr: 0.8187
#> Update Epoch 3/5 | CVs: 32 | QE: 0.3796 | so_lr: 0.4493
#> Update Epoch 4/5 | CVs: 32 | QE: 0.3452 | so_lr: 0.1653
#> Update Epoch 5/5 | CVs: 32 | QE: 0.3151 | so_lr: 0.0408