itecture by starting with a useful application like a search engine or conversational agent, and extending its common sense abilities over time. But Ben Kuipers worried that choosing too specific an application would lead to what happened to most previous projects-- someone discovers some set of ad hoc trick

itecture by starting with a useful application like a search engine or conversational agent, and extending its common sense abilities over time. But Ben Kuipers worried that choosing too specific an application would lead to what happened to most previous projects— someone discovers some set of ad hoc tricks

spacetime is necessary for achieving humanlike AGI using feasible resources. As an argument against the necessity of this sort of prior knowledge, Ben Kuipers and his colleagues ISNIK 21 have shown that an AI system can learn via experience that its perceptual stream comes from a world with three, rather

urations that carry out sensorimotor and categorization tasks. The emergence of higher-level cognition from this approach seems rather unlikely. • Ben Kuipers and his colleagues INIK07, MKOS, NIKOalhave pursued an extremely innovative research program which combines qualitative reasoning and reinforcement

r the various cognitively-relevant properties of the environment from looking at data about the ev- eryday world. We agree in principle, and in fact Ben Kuipers and his colleagues have done some interesting work in this direction, showing that learning algorithms can infer some basics about the structure of

igurations that carry out sensorimotor and categorization tasks. The emergence of higher-level cognition from this approach seems rather unlikely. e Ben Kuipers and his colleagues [MK07, MIX08, MIXO09]have pursued an extremely innovative research program which combines qualitative reasoning and reinforcement

er the various cognitively-relevant properties of the environment from looking at data about the ev- eryday world. We agree in principle, and in fact Ben Kuipers and his colleagues have done some interesting work in this direction, showing that learning algorithms can infer some basics about the structure of s

itecture by starting with a useful application like a search engine or conversational agent, and extending its common sense abilities over time. But Ben Kuipers worried that choosing too specific an application would lead to what happened to most previous projects-- someone discovers some set of ad hoc trick