V is constrained to vary along a limited set of dimensions.

Several additional assumptions were suggested to make the ICA problem identifiable.

Identifying interpretable patterns in multi-electrode recordings is a longstanding and increasingly pressing challenge in neuroscience.

One major concern of AM is the heavier memory overhead compared with the fixed-point iteration.

To do so, we introduce a dataset of simulated touch and vision signals from the interaction between a robotic hand and a large array of 3D objects.