Speaker: Dana Wilkinson
In a variety of domains it is desirable to learn a representation of
an environment defined by a stream of sensori-motor experience. In
many cases such a representation is necessary as the observational
data is too plentiful to be stored in a computationally feasible
way. In other words, the primary feature of a learned representation
is that it must be compact, summarizing information in a way that
alleviates storage and retrieval demands.
This admits a new way of phrasing the problem: as a variation of
dimensionality reduction. There are a variety of well-studied
algorithms for the dimensionality reduction problem. We argue that any
of these can be useful for learning compact representations as long as
additional constraints to the problem are respected, namely that the
resulting representation is useful in the context of the actions which
generated the observations.
Here, we formalize the problem of learning a subjective
representation, clearly articulating solution features that are
necessary for a learned representation to be ``useful''; the actions
must correspond to simple and consistent transformations in the
learned representation. Further, we briefly present a possible
solution to the newly defined problem and demonstrate it's
effectiveness for reasoning, planning and localization.