Abstract: The field of artificial intelligence (AI) has seen several proposals for modeling intelligent computation. Two of the most popular ones are (1) neural – which is inspired by the structure of our brain and consists of millions of nodes resembling neurons connected in a network, and (2) symbolic – which uses the formalism of logic to make inferences from known facts. While deep neural models have revolutionized the field of AI in modern times, an emerging body of work combines neural models with symbolic computation to achieve the best of both worlds. In this introductory tutorial, we briefly present some of this literature in the context of (1) augmenting neural models by incorporating additional symbolic knowledge, (2) designing neural models for solving symbolic reasoning problems, and, (3) neuro-symbolic architectures for solving perceptual-reasoning tasks.
Bio: Yatin is a PhD scholar in the area of Machine Learning and Artificial Intelligence, guided by Mausam and Parag Singla at Computer Science and Engineering Department, Indian Institute of Technology Delhi.
Prior to joining the PhD program in 2017, he worked in the quantitative finance industry for 10 years.
During the last five years of his professional stint, he was busy making high frequency trading strategies at Estee Advisors, trading primarily in stock, index and currency options.
Yatin started my career in 2007 with the Equity Quantitative Analytics team at Lehman Brothers, which was eventually bought by Nomura after its bankruptcy in 2008.
He did his graduation in Mathematics and Computing, a five year integrated M.Tech programme offered by Mathematics Department at IIT Delhi.