Double Deep Machine Learning

Very important breakthroughs in data-centric machine learning algorithms led to impressive performance in ‘transactional’ point applications such as detecting anger in speech, alerts from a Face Recognition system, or EKG interpretation. Nontransactional applications, e.g. medical diagnosis beyond the EKG results, require AI algorithms that integrate deeper and broader knowledge in their problem-solving capabilities, e.g. integrating knowledge about anatomy and physiology of the heart with EKG results and additional patient’s findings. Similarly, for military aerial interpretation, where knowledge about enemy doctrines on force composition and spread helps immensely in situation assessment beyond image recognition of individual objects. An initiative is proposed to build Wikipedia for Smart Machines, meaning target readers are not human, but rather smart machines. Named ReKopedia, the goal is to develop methodologies, tools, and automatic algorithms to convert humanity knowledge that we all learn in schools, universities and during our professional life into Reusable Knowledge structures that smart machines can use in their inference algorithms. Ideally, ReKopedia would be an open source shared knowledge repository similar to the well-known shared open source software code repositories. The Double Deep Learning approach advocates integrating data-centric machine self-learning techniques with machineteaching techniques to leverage the power of both and overcome their corresponding limitations. For illustration, an outline of a $15M project is described to produce ReKo knowledge modules for medical diagnosis of about 1,000 disorders. AI applications that are based solely on data-centric machine learning algorithms are typically point solutions for transactional tasks that do not lend themselves to automatic generalization beyond the scope of the data sets they are based on. Today’s AI industry is fragmented, and we are not establishing broad and deep enough foundations that will enable us to build higher level ‘generic’, ‘universal’ intelligence, let alone ‘super-intelligence’. We must find ways to create synergies between these fragments and connect them with external knowledge sources, if we wish to scale faster the AI industry. Examples in the article are based onor inspired byreal-life non-transactional AI systems I deployed over decades of AI career that benefit hundreds of millions of people around the globe. We are now in the second AI ‘spring’ after a long AI ‘winter’. To avoid sliding again into an AI winter, it is essential that we rebalance the roles of data and knowledge. Data is important but knowledgedeep and commonsenseare equally important.