Follow me, now I'll explain what I mean. In this article: Google MUM, artificial intelligence and complex queries Google MUM: multitasking and multilingual learning The effects of Google MUM on content marketing Will computers steal our jobs? Google MUM, artificial intelligence and complex queries Let's start from the beginning. MUM is an acronym that stands for Multitask Unified Model. (“Multitask” is a key word, but we'll get back to it shortly). In its official blog , Google called MUM “the new milestone in artificial intelligence to better understand information.
” Compared to its big brother BERT, Google MUM has germany telegram data dramatically multiplied the neural connections of the algorithm. To show you what I mean, the new algorithm is estimated to be about 1000 times more powerful than the previous one. And if Google BERT had already made gigantic steps forward in the field of artificial intelligence and machine learning, Google MUM is definitely many steps above. Ok, but what does this mean in practice? Google MUM moves very quickly between the entities of the Knowledge Graph.
And thanks to this speed it is able to process answers to complex queries to provide directly to the user. Google itself provides us with an example of what it means to answer complex queries. Until today, if we wanted to go for a hike on Mount Fuji we could prepare ourselves through a series of targeted searches, therefore asking Google (for example): The average temperature in a certain season The difficulty of the paths But we couldn’t ask Google for a purely human opinion, like: “I’ve already done other climbs.