Integrating Learning and Inference using Cognitive
Both learning and reasoning are important aspects of intelligence. However they are rarely integrated within a single agent. Here it is suggested that im-precise learning and crisp reasoning may be coher-ently combined via the cognitive context. The identification of the current context is done using an imprecise learning mechanism, whilst the con-tents of a context are crisp models that may be use-fully reasoned about. This also helps deal with situations of logical under- and over-determination because the scope of the context can be adjusted to include more or less knowledge into the reasoning process. An example model is exhibited where an agent learns and acts in an artificial stock market.