Twenty Significant Problems in AI Research, with Potential Solutions via the SP Theory of Intelligence and Its Realisation in the SP Computer Model

This paper highlights 20 significant problems in AI research, with potential solutions via the SP Theory of Intelligence (SPTI) and its realisation Computer Model. With other evidence referenced paper, this is strong support SPTI as a promising foundation for development human-level broad AI, aka artificial general intelligence. The include: tendency deep neural networks to make major errors recognition; need coherent account generalisation, over- under-generalisation, minimising corrupting effect ‘dirty data’; how achieve one-trial learning; transfer transparency representation processing knowledge; eliminate problem catastrophic forgetting. In addition promise AGI, has study human learning, perception, cognition. And it mathematics, logic, computing.