As the world is advancing, the new age technologies are bound to stay at their ever-newest form where being considered as new is not enough, rather they need to evolve to update themselves as the demands grow. A number of experts, in space of AI, believe that it needs to change with time. There are various approaches to artificial intelligence, some at the forefront while others under some layers (like neural networks and symbolic AI). The AI innovations certainly don't mean to bring about change in the whole arena. Small combinations and blending of different AI approaches could also bring about invention, innovation and better implementation into the fast-paced world. Moreover, a number of organizations and research centers are into innovating ways and methods AI is being implemented and amid this MIT-IBM Watson is calling for change for the greater good.

According to David Cox, director of the MIT-IBM Watson A.I. Lab in Cambridge, MA, AI needs to change, however, his suggestion for that change is a currently obscure term called "neuro-symbolic AI," that could become one of those phrases the world is intimately acquainted with by the time the 2020s come to an end.

As noted by Digital Trends, Neuro-symbolic AI is not a totally new way of doing AI rather it is a combination of two existing approaches to building thinking machines; ones which were once pitted against each as mortal enemies. Where on one hand the "symbolic" part of the name refers to the first mainstream approach to creating artificial intelligence, the "neuro" part of neuro-symbolic AI, on the other hand, refers to deep learning neural networks.

The neural networks work differently to symbolic AI because they're data-driven. To explain something to a symbolic AI system one needs to explicitly provide with every bit of information it needs to be able to make a correct identification while to train a neural network to do such task, one needs to simply show it thousands of pictures of the object in question and once it gets smart enough, it will be able to recognize that object and make up its own similar objects. There is a possibility that such objects never actually existed in the real world.

However, the idea behind neuro-symbolic AI is to bring together these approaches to combine both learning and logic. It is obvious that neural networks will help make symbolic AI systems smarter by enabling it to simplify the world into symbols, rather than relying on human programmers to do it for them. Moreover, symbolic AI algorithms will help translate common sense reasoning and domain knowledge into deep learning. This could subsequently lead to significant advances in AI systems tackling complex tasks, relating to everything from self-driving cars to NLP while requiring much less data for training.

David Cox said, "Neural networks and symbolic ideas are really wonderfully complementary to each other. Because neural networks give you the answers for getting from the messiness of the real world to a symbolic representation of the world, finding all the correlations within images. Once you've got that symbolic representation, you can do some pretty magical things in terms of reasoning."