in Non Technical

Relational reasoning as the basis for human intelligence

A blog and research paper from Google Deepmind brought our attention to the concept of relational reasoning. As humans we have the innate ability to connect the dots or plot a narrative from piece of information to make a decision either to run a search, make a purchase or predict the outcome of a movie. Artificial agents are yet to attain the creative human ability to connect entities together via a narrative exercise that leads to human action. A few days ago my wife told me an interesting story. She ran into a friend at the entrance of a shop, he’s just grabbed himself a drink and some popcorn. She asked him where he was headed and he said to West India Quay. It is a place in east London which boasts of a handful of restaurants, bars and a cinema. She pieced together the the popcorn, drinks and West Indian Quay and asked him if he was headed to the cinema? He was quite surprised and affirmed he was headed to the cinema. She then predicted or stated that he owns a monthly Cineworld membership. He was quite shocked and nodded to having a monthly Cineworld membership. She told me, her experience of buying popcorn, drink and preferring the West India Quay Cineworld with her Meetup movie mates, assisted her in creating a relationship and narrative from the little information she received to correctly predict the intention of her friend. This is relational reasoning at work as the deepmind team clearly mentioned “ We carve our world into relations between things. And we understand how the world works through our capacity to draw logical conclusions about how these different things – such as physical objects, sentences, or even abstract ideas – are related to one another.”

As humans, in most cases, we use narratives to connect things before taking an action. Our intelligence as humans is somewhat enhanced, when we creatively construct a story or find a connection between things and anticipate or predict their impact. Formisano, Omodeo and Simeoni worked on a piece titled ‘A  graphical approach to relational reasoning.’ Whilst their article argues that relational reasoning exceeds dyadic relations (relationship between two people connected through romance, love, interest, work and so on). They projected the representation of relations based on labeled graph. It is a maths heavy piece that looked into three uses of graph representation of relations with the first being translating first-order specifications into the calculus of relations. The second was aimed at inferring equalities within calculus and the third explores the specialised framework of set theory. The authors opting for a graphical representation of relational reasoning is because they believe it presents the moving pictures of thought. The idea is not to delve into the mathematical and graphical workings in this blog but to make you understand that graph theory and implementation has a place in relational reasoning.

Narratives are important aspects of our life and they are central to human intelligence. We utilise narratives to create a relational reasoning pattern that guides us into making the best search on Google, buy the most relevant product from the eCommerce store or prevent us from shoplifting as we connect the dots that it could lead to a criminal record which could make getting a job slightly difficult.

I have been particularly inspired by the work of Patrick Henry Winston on the work ‘The Strong Story Hypothesis and the directed perception hypothesis.’ We will explore these theories in more depth in future blogs but it is important to note that the directed perception hypothesis has a connection with the concept of relational reasoning. Patrick Winston enthuses that this hypothesis is crucial to the application of common sense. It involves the application of our perceptual senses on real and imagined event. The example in the introductory section of the deepmind blog on relational reasoning is a representation of the direct perception hypothesis as the example goes thus ‘… a child who runs ahead of her ball to prevent it from rolling into a stream.’ The ball rolling is a real event and going into a stream is an imagined event. The child had to create a story or connection of a real event (ball rolling) and an imagined one (rolling into a stream) for an intelligent action (running after the ball).

Write a Comment

Comment