Breaking Barriers in Numerical AI

This 16-minute conference talk from DevConf.IN 2025 explores how Large Quantitative Models (LQMs) are transforming numerical AI applications. Discover how these innovative models combine Variational AutoEncoders (VAEs) and Generative Adversarial Networks (GANs) to overcome challenges in quantitative datasets, including high variability, limited historical data, and complex relationships. Learn about LQMs' versatile applications beyond their original financial forecasting purpose, extending to IoT sensor predictions and healthcare patient outcome simulations. Gain insights into the architecture of LQMs and leave with practical knowledge on how to experiment with these models using open-source tools, opening new possibilities for quantitative analysis across various domains.