Multitask Learning

By Aviv Navon and Shai Meital @ Aiola.com Introduction and motivation Natural Language Processing (NLP) is one of the most researched and used practices in applied data science, thanks to many fundamental breakthroughs and broad applicability across every industry with natural language communication. Four of the most fundamental tasks of NLP are (1)…

With inputs from Rutvik Vijjali, Mithun TM Introduction Most people have a sense of knowing the similarity between items, i.e, given an item say “Veg Burger’’, we can relate it to a list of similar items, starting from burgers, to pizzas, fries and so on. Can machines do this? Possibly. This…

Acquring and assimilating knowledge from multiple domains is a central tenet of human intelligence. For example, when a newborn baby learns to walk it acquires and assimilates general motor skills. Later in life, it uses and builds on these skills to perform more complex tasks such as playing soccer. In…

Is Multi-Tasking Bad for the Brain? — That evening I was at a friend’s house for a wedding ceremony. The ceremony was held on a wide lawn of the house. All the guests were engaged in gossip when suddenly we heard the voice of broken glass. We found out that a tray full of water glasses was…

Towards building a “Generalist ” model Multi-task Learning (MTL) is a collection of techniques intended to learn multiple tasks simultaneously instead of learning them separately. The motivation behind MTL is to create a “Generalist” model that can solve multiple tasks rather than creating multiple “Specialist” models that are trained to…

Classification involves predicting if something belongs to a class among a group of classes. First of all, we define the multi-task and multi-label classification terms. A classification task with just two classes is called binary classification (e.g. “red” or “blue”, “success” or “failure”, etc), while for any larger number of…

Using neural networks and with the help of high level tensor libraries we can build models that can handle classification, regression and other tasks easily. When I did coursera specialization on deep learning, watching a video on multi-label learning by Andrew Ng I quickly set up my mind to try…