The third edition of the Indian Symposium on Machine Learning (IndoML), hosted at the Indian Institute of Technology Gandhinagar (IITGN) in collaboration with IIT Kharagpur, concluded on December 17, 2022. IndoML 2022 brought together researchers and practitioners from diverse domains to share their knowledge and experience on the significance of the intersection of Artificial Intelligence (AI) and Machine Learning (ML) with other fields of research.

More than 300 participants from 49 academic institutions, such as IIT Kharagpur, IIT Bombay, IIT Delhi, and several state colleges, such as DAIICT, Jadavpur University, Ahmedabad University, and industry representatives from nine companies, such as Rakuten, Merlyn Mind, HP Enterprises participated in IndoML 2022. The symposium included 17 talks, tutorials, and poster presentations by eminent speakers, researchers, and industry personnel from across the globe, including GeorgiaTech, Rice University, Northeastern University, UMass Amherst, University of Michigan, the University of Texas at Dallas, NIST, IISc Bangalore, IIT Kharagpur, IIT Bombay, IIT Jodhpur, Google, Amazon, Merlyn Mind, Accenture, Hewlett Packard, and so on.

The three-day symposium, held from December 15 to 17, covered a variety of topics, challenges, and promising solutions from the field of AI and ML, such as using ML techniques for predicting chemical properties of crystalline materials, modelling climate change, robotics, the intersection of ML and economics, sustainable solutions to the growing problem of AI carbon emissions, and so on. Talks on AI ethics and fairness and using AI to interpret political and ideological biases in US presidential elections generated significant interest among the participants.

Speaking about “Human-Centered Artificial Intelligence – A Crowd Computing Perspective”, Prof Ujjal Gadiraju, Assistant Professor at Delft University of Technology, Netherlands, said, “Crowd Computing offers a promising means to overcome fundamental challenges in computation and interaction. It can herald a new generation of Human-centered AI systems”.

Another expert speaker, Prof Anshumali Shrivastava, Associate Professor at Rice University and Founder of ThirdAI, stressed on the promise of Dynamic Sparsity algorithm and said, “The hurdle in Universal AI is the scale of models. Simple architectures are fine, but we cannot ignore model size, as large models perform better. Dynamic Sparsity is the idea of using different subsets of parameters in the whole network for different inputs. This can allow training of billion parameter models on CPUs significantly faster.”

IndoML 2022 also hosted a Datathon for students as well as early career professionals. The event received more than 600 submissions from 107 teams. Selected top teams were invited to IndoML 2022 to present their work before the expert panel of ML researchers and practitioners.

Several leading national and international tech companies, including ACM, Merlyn Mind, Hewlett Packard, Deepmind, Accenture, Swiggy, Nutanix, Netweb Technologies, Google, and the Artificial Intelligence Journal had partnered with IITGN for IndoML 2022. The next edition of IndoML will be hosted at IIT Bombay.