Cracking the mystery of how Artificial Intelligence works

K V Kurmanath Hyderabad | Updated on September 10, 2019 Published on September 10, 2019

Artificial Intelligence models and programmes mimic the functioning of human brains, helping machines learn make decisions in a more-like manner. But how exactly they arrive at decisions is unknown.

In order to understand this, a group of researchers at the Indian Institute of Technology (IIT-Hyderabad) have developed a method by which the inner workings of Artificial Intelligence models can be understood in terms of causal attributes.

This finding assumes significance in the wake of emergence of regulations such as General Data Protection Regulation (GDPR) that requires organisations to explain the decisions made by machine learning methods.

Modern Artificial Neural Networks, also called Deep Learning (DL), have increased tremendously in complexity such that machines can train themselves to process and learn from data that has been fed to them. They almost match human performance in many tasks.

“Transparency and understandability of the workings of DL models are gaining importance as discussions around the ethics of Artificial intelligence grows,” Vineeth N. Balasubramanian, Associate Professor (Department of Computer Science and Engineering) at IIT- Hyderabad, said.

He along with his students Aditya Chattopadhyay, Piyushi Manupriya, and Anirban Sarkar have worked on the project. Their work has recently been published in the Proceedings of 36th International Conference on Machine Learning.

“There are challenges to be met as the achievements in the field of AI and machine learning have wowed everyone,” he said.

“A key bottleneck in accepting such Deep Learning models in real-life applications, especially risk-sensitive ones, is the ‘interpretability problem,” he said.

“The DL models, because of their complexity and multiple layers, become virtual black boxes that cannot be deciphered easily.

This makes troubleshooting difficult, if not impossible,” Balasubramanian explained.

The IIT Hyderabad team approached this problem with ANN architectures using causal inference with what is known in the field as a ‘Structural Causal Model.’

“We have proposed a new method to compute the Average Causal Effect of an input neuron on an output neuron. It is important to understand which input parameter is ‘causally’ responsible for a given output,” he pointed out.

For example in the field of medicine, how does one know which patient attribute was causally responsible for the heart attack? “Our (IIT Hyderabad researchers’) method provides a tool to analyse such causal effects,” he added

Published on September 10, 2019

A letter from the Editor

Dear Readers,

The coronavirus crisis has changed the world completely in the last few months. All of us have been locked into our homes, economic activity has come to a near standstill. Everyone has been impacted.

Including your favourite business and financial newspaper. Our printing and distribution chains have been severely disrupted across the country, leaving readers without access to newspapers. Newspaper delivery agents have also been unable to service their customers because of multiple restrictions.

In these difficult times, we, at BusinessLine have been working continuously every day so that you are informed about all the developments – whether on the pandemic, on policy responses, or the impact on the world of business and finance. Our team has been working round the clock to keep track of developments so that you – the reader – gets accurate information and actionable insights so that you can protect your jobs, businesses, finances and investments.

We are trying our best to ensure the newspaper reaches your hands every day. We have also ensured that even if your paper is not delivered, you can access BusinessLine in the e-paper format – just as it appears in print. Our website and apps too, are updated every minute, so that you can access the information you want anywhere, anytime.

But all this comes at a heavy cost. As you are aware, the lockdowns have wiped out almost all our entire revenue stream. Sustaining our quality journalism has become extremely challenging. That we have managed so far is thanks to your support. I thank all our subscribers – print and digital – for your support.

I appeal to all or readers to help us navigate these challenging times and help sustain one of the truly independent and credible voices in the world of Indian journalism. Doing so is easy. You can help us enormously simply by subscribing to our digital or e-paper editions. We offer several affordable subscription plans for our website, which includes Portfolio, our investment advisory section that offers rich investment advice from our highly qualified, in-house Research Bureau, the only such team in the Indian newspaper industry.

A little help from you can make a huge difference to the cause of quality journalism!


Support Quality Journalism
This article is closed for comments.
Please Email the Editor
You have read 1 out of 3 free articles for this week. For full access, please subscribe and get unlimited access to all sections.