Co-adaptive learning systems involve decomposition by definition: concurrent learning algorithms working on different aspects of a problem, game, or simulation imply (explicitly or implicitly) a division in representation of some kind. For many such systems, this partitioning is performed explicitly and *a priori* (e.g., traditional compositional co-evolutionary approaches). The representational issues surrounding how to choose to decompose a problem for the successful application of co-adaptive learning approaches are complex and often quite counter intuitive. For example, the property of linear separability is neither a sufficient nor necessary condition for success.

Though some of these issues have been addressed for specific classes of algorithms, such as the aforementioned compositional approach, we believe many of them are general to a variety of co-adaptive methods. Our approach is to explore decompositional issues from both theoretical and empirical points of view — to answer such questions as: What properties of the problem affect representational choices with respect to decomposition in co-adaptive learning algorithms? How do other algorithmic design choices relate to the question of decomposition? What are the trade offs between optimal decompositions and intuitive decompositions?