About

Sidak Pal Singh is a highly accomplished PhD student in machine learning at ETH Zurich and MPI Tubingen. With over 2.68 years of relevant experience in the field of AI and ML, Sidak has worked with some of the most prestigious organizations in the industry, including Facebook AI Research, EPFL, and IST Austria. Sidak's expertise lies in developing efficient second-order methods for model compression and in working with latent variable models for text representation and generation. He has published several research papers in renowned journals and has presented his work at various international conferences. Currently working as a Research Assistant at the Institute of Science and Technology Austria, Sidak is focused on his Master thesis on Efficient second-order methods for model compression. He has also worked as a Research Intern at Facebook AI Research, where he developed GLOSS: Generative Latent Optimization of Sentence Representations. In his previous role as a Research Scholar at MLO Lab, EPFL, Sidak proposed a novel method to represent each entity as a distribution over its contexts, which led to the development of the Context Mover's Distance (CMD) for measuring any kind of distance between entities. He also extended this method to form a representation of the composition of entities using Wasserstein Barycenter, which proved to be a competitive method for use in sentence similarity tasks. Sidak holds a Doctor of Philosophy - PhD from ETH Zurich, a Master of Science - MS in Data Science from EPFL, and a Bachelor of Technology (BTech) in Computer Science from IIT Roorkee. He completed his schooling with an impressive 95.8% from Sacred Heart Senior Secondary School. With a strong tech stack that includes Research Scientist, AI, and ML, Sidak is a highly skilled professional who is passionate about his work. He is known for his ability to solve complex problems, think outside the box, and come up with innovative solutions.