As the new IBM Research India lab leader, Gargi B Dasgupta’s priority for 2019 is to spearhead specialised research in the areas of Enterprise AI and Blockchain. She stresses that her focus will be to reimagine the state of today’s businesses by building on the future of computing. “My goal is to embed IBM India’s research in every aspect of the India/South Asia innovation ecosystem,” Dasgupta, director, IBM Research-India & CTO, IBM India/South Asia, tells Sudhir Chowdhary in a recent interview. Excerpts:
What are the key sectors IBM Research India lab is focused on?
IBM Research India lab is focused on building trusted Artificial Intelligence (AI) on our scalable hybrid cloud platforms and applying it to various domains including retail and agritech in India. For retail, we have developed an AI technology and solutions that allow brick-and-mortar retailers to leverage their core strengths such as data to help them move into the new era of Retail 3.0. Our work with Bestseller leveraging Watson helped it analyse how it may perform in stores. Today, Bestseller is embarking on a journey to become a fully AI-driven enterprise. Bestseller plans to use AI to determine the right assortment plan for each store, predict the next best product to incorporate into its mix, and improve the efficiency of its supply chain.
In agritech, we are combining multiple global satellite-based information sources to compute actionable agronomic insights (such as crop health stress alert, water stress, pest/disease risk forecast, etc.) for farmers at sub-acre level. Combining this with the weather data from The Weather Company and Watson we have launched the Watson Decision platform for agriculture. We are tackling challenges around crop insurance by developing AI solutions customised for the insurance sector. We are also developing solutions to address farmer credit through cognitive credit farm scoring application. This application, using basic farm details, would be helpful in determining how much yield and profit (based on market prices) a farmer would have made in the past few crop seasons. This information can be used to determine a more accurate and just estimation of a farmer’s credit-worthiness.
Any specific projects at the ground level?
Yes, we recently announced our collaboration with the Karnataka Agricultural Prices Commission (KAPC) to develop an advanced price forecasting system—a dashboard leveraging IBM’s Watson Decision Platform for Agriculture. It will help to assess the acreage on a real-time basis and predict the market price trends for at least a fortnight and the production pattern of tomatoes. The price forecasting mechanism for Karnataka is the first of its kind in the country. It is being launched in the tomato-growing districts of Kolar, Chikkaballapur and Belgavi and two key maize-producing districts of Davangere and Haveri.
How important is IBM Research India lab for IBM labs globally?
IBM India Research lab plays a significant role in the infusion of AI within enterprises through its multi-cloud strategy. Blockchain is also a key focus for us. Our country’s diversity makes it ideal training ground for AI and Blockchain research and develop proof of concept to apply in real-time.
What are your areas of research in AI?
IBM Research has been exploring AI and machine learning for decades. Our research agenda is anchored in three main areas: Advancing the capabilities of core AI technologies in areas that humans excel at, such as language, vision, and reasoning; embedding and operationalising these AI capabilities into businesses with a platform that is secure; and pushing the frontiers of the underlying infrastructure to optimise it for AI services. In India, IIT Delhi and IIT Bombay are part of IBM AI Horizons Network to advance experiments designed to accelerate the application of AI and related fields.
What are the skills do you see researchers need in the AI and Blockchain era?
The sought-after skills for AI will be machine learning, conversational AI, natural language processing along with Blockchain crypto and distributed systems and algorithms.
