Take away lessons

NEF-based representation: Distributed representation of mathematical constructs using nonlinear - spike-based - encoding and linear decoding.

Orthogonal and over-complete basis for representation. In the canonical description of vector representation, the spanning vectors are orthogonal. If we relaxed this constraint, such that these vectors were not necessarily independent, we might end up with an over-complete basis. With an over-complete representation, the number of the basis vectors is larger than the dimensionality of the input and it is, therefore, considered less efficient

NEF-based transformation: The transformation of one representation to another through weighted synaptic connections between two neuron ensembles.

NEF-based dynamics: The manifestation of a dynamical system from recurrently connected neuron ensembles.