Grounding Symbols: Labelling and Resolving Pronoun Resolution with fLIF Neurons

If a system can represent knowledge symbolically, and ground those symbols in an environment, then it has access to a vast range of data from that environment. The system described in this paper acts in a simple virtual world. It is implemented solely in fatiguing Leaky Integrate and Fire neurons; views the environment; processes natural language commands; plans; and acts. Visual representations are labeled, using a Hebbian learning rule, thus gaining associations with symbols. The labelling is done using simultaneous presentation of the label and a corresponding visual item. These grounded symbols can be useful in reference resolution. Both experiments perform perfectly on all tests.