There is no doubt that machine learning is one of the major driving forces behind most of the advanced techs and gadgets we have today. Whether it is your smart home device or that newly bought self-driving car yours, ML is playing a vital role not only advancing gadgets but is also changing the way people interact with machines. No doubt, it is one of the hottest technologies in the world.
However, people almost forget that there is something called Machine Teaching that also plays a significant role in all the ML use cases.
Subscribe to our Newsletter
Join our editors every weekday evening as they steer you through the most significant news of the day, introduce you to fresh perspectives, and provide unexpected moments of joy
The Teacher’s Perspective
We all have heard a lot about ML, which is also seen as a subset of artificial intelligence. And if we try to understand it by its literal definition, it is the ability of algorithms and statistical models of machines or systems to automatically learn and improve from experience without being programmed or without using any explicit instructions.
When we talk about this sought after technology, artificial intelligence, people tend to think about things like performance modelling, learning algorithm optimisation, network architecture, all the tasks it can perform. However, this is just through the perspective of a learner.
But what it would be when we look at the entire scenarios through a teacher’s perspective? Would it still be the same things?
Microsoft has the answer. While the rest of the world has been always focused on the learner side, tech giant Microsoft has taken the teacher’s perspective and explored the possibilities where a machine could be able to do all the tasks that a human can do — whether it’s about thinking, finding solutions, etc.
However, there was a bottle that stands as a hurdle for machine learning to accomplish these goals. But when you go about taking a deeper look for a solution to it, you realise the bottle is not the technology but the teacher — the data. And this where Machine Teaching comes into the scenario.
What Exactly Is Machine Teaching?
Machine teaching is an approach where human expertise and abilities to find solutions to problems are used in order to help machine learning models find important hints about how to find a solution faster. To be more technical about it, Machine teaching designs the optimal training data to drive the learning algorithm to a target model.
Unlike the traditional approach of teaching machine which about feeding the machine or the system with lots of available data, machine teaching takes a different route to go about it — it not only feeds the data but also let know where to look.
For example, when a teacher teaches a student how to recognise a bicycle, the teacher tells the students about all the features and characteristics of the bicycle and later tests him/her with other things. And if the students say the Motorbike is the bicycle, the teacher doesn’t say it is wrong; rather, the teachers correct the student by telling him/her about the differences.
And this is the same loop that machine teaching uses. Rather than extracting data and digging insights from it to train a model, people’s expertise could also be a significant driving force in teaching machines.
Technology is on an evolving spree and year by year things are just getting better and bigger. The same has happened with machine learning. According to Microsoft, who has been researching and working on Machine Teaching for a decade now, even AI struggles and sometimes fail to learn things by itself, but when people start to guide the machine to do and learn things that we already know, things are going to significantly different.
The machine is not only a whole new approach to machine learning but it’s an approach to empower people to make sophisticated use of AI. It doesn’t matter whether you are a developer or an SME with limited knowledge, machine learning makes things easier — one can impart abstract concepts to an intelligent system, and it would perform the machine learning mechanics in the background.