Skip to yearly menu bar Skip to main content


Poster

Neurocoder: General-Purpose Computation Using Stored Neural Programs

Hung Le · Svetha Venkatesh

Hall E #419

Keywords: [ DL: Sequential Models, Time series ] [ DL: Attention Mechanisms ] [ DL: Recurrent Networks ] [ Deep Learning ]


Abstract:

Artificial Neural Networks are functionally equivalent to special-purpose computers. Their inter-neuronal connection weights represent the learnt Neural Program that instructs the networks on how to compute the data. However, without storing Neural Programs, they are restricted to only one, overwriting learnt programs when trained on new data. Here we design Neurocoder, a new class of general-purpose neural networks in which the neural network “codes” itself in a data-responsive way by composing relevant programs from a set of shareable, modular programs stored in external memory. This time, a Neural Program is efficiently treated as data in memory. Integrating Neurocoder into current neural architectures, we demonstrate new capacity to learn modular programs, reuse simple programs to build complex ones, handle pattern shifts and remember old programs as new ones are learnt, and show substantial performance improvement in solving object recognition, playing video games and continual learning tasks.

Chat is not available.