Speaker "Ganes Kesari Balakesari" Details Back



Latest trends in Visualising Machine Learning: A case-study driven approach to humanising the intelligence


These are exciting times for the discipline of data science. While we make stellar advances in AI and further the cognitive capability of machines, there is a rising disenchantment & distrust in the technology. One of the key factors at the root of this growing dichotomy is the lack of understanding in machine learning and deepening opaqueness of the algorithms. This talk will highlight this challenge by using industry case studies to illustrate practical roadblocks faced by enterprises with the adoption of machine learning and AI initiatives. The solution framework that can be leveraged to tackle this challenge head-on will be presented by showcasing some of the relevant, upcoming trends in Information Design, Data Visualisation and Machine learning that are particularly useful in this perspective. Through live examples, the successful implementation of such techniques across domains and business use cases will be presented.


Ganes co-founded Gramener and he currently spearheads innovation in AI, machine learning & information design across the organization, and is based out of the Princeton, NJ office. With over 15 years of experience in building and scaling organisations around technology, data and management, Ganes advises business enterprises on setting up and deriving value from data science initiatives. He speaks at data science conferences, startup summits and delivers lectures in data science at leading universities like the Indian School of Business (ISB, Hyderabad), National Institute of Design (NID, Bangalore). In his career, he has played a variety of strategic leadership roles managing large, complex software engagements at HCL Technologies, Cognizant and Birlasoft. Ganes holds a Bachelors degree in Engineering from CECRI and MBA (full-time) from FMS, Delhi University, India. A sample set of video clips from his earlier presentations can be found here: