This paper proposes a continuous-time machine learning model that learns the chronological relationships and the intervals between events, stores and organises the learnt knowledge in different levels of abstraction in a network, and makes predictions about future events. The acquired knowledge is represented in a categorisation-like manner, in which events are categorised into categories of different levels. This inherently facilitates the categorisation of static items and leads to a general approach to both spatial and temporal perception. The paper presents the approach and a demonstration showing how it works.
展开▼
