Sudarshan Babu
PhD student, |
I am a PhD student at Toyota Technological Institute affiliated to the University of Chicago. I am fortunate to be advised by Michael Maire and Greg Shakhnarovich . I am also grateful for Rana Hanocka's advise on all things 3D.
Broadly, I am interested in designing architectures and training routines to enable networks to adapt faster on tasks with limited data.
I have designed several types of hypernetworks for rapidly adapting on tasks with limited data. Please check out my recent work in using hypernetworks to accelerate 3D generation under data constraints: HyperFields.
I am interested in accelerated training for 3D tasks such as reconstruction, generation, segmentation.
Recently, intersted in applying 3D tools developed in computer vision to problems in computational biology.
I am currently looking for academic and industry positions. [Curriculum Vitae]HyperFields: Towards Zero-Shot Generation of NeRFs from Text [PDF] [Website] [Hugging Face]
Sudarshan Babu*, Richard Liu*, Avery Zhou*, Michael Maire, Greg Shakhnarovich, Rana Hannocka
Under Review
Online Meta-Learning via Learning with Layer-Distributed Memory [PDF]
Sudarshan Babu, Pedro Savarese, Michael Maire
NeurIPS 2021
Domain-independent Dominance of Adaptive Methods [PDF]
Pedro Savarese, David Mcallester, Sudarshan Babu, Michael Maire
CVPR 2021
HyperNetwork Designs for Improved Classification and Robust Meta-Learning [PDF]
Sudarshan Babu, Pedro Savarese, Michael Maire
2020
HyperSegmentationFields: Towards Zero-shot Generation of Segmentation Fields
3D Segmentation; HyperNetworks for generating segmentation fields
Controllable Novel-View Synthesis via Hyper Codes
Controlled semi generative 3D model via learnt latent codes; HyperNetworks map codes to NeRFs
Nvidia-Research Scientist Intern
Designing systems to train non-stationary long tail distributions
Amazon-Research Scientist Intern
Learning Efficient Curriculum for Training Neural Networks
co-organized the TTIC Student Workshop 2020.
Teaching Assistant for the Statistical Machine Learning for Fall 2020.