Sandesh Adhikary
Resume
(Click for PDF)Education
2019-Present
PhD Student, Computer Science and Engineering, University of Washington
2017-2019
PhD Student, Computational Science and Engineering, Georgia Tech.
2011-2015
Bachelors of Arts, Physics, Reed College
Research
2022-Present
Geometric Representations for Reinforcement Learning
Developing reinforcement learning algorithms that exploit geometric structures in decision processes. Most recently, improved performance of behavioral-distance based reinforcement learning algorithms through locality-preserving Laplacian eigenmaps.
Papers
Developing reinforcement learning algorithms that exploit geometric structures in decision processes. Most recently, improved performance of behavioral-distance based reinforcement learning algorithms through locality-preserving Laplacian eigenmaps.
Papers
Adhikary, S., Li, A., and Boots, B., (2024). BeigeMaps: Behavioral Eigenmaps for Reinforcement Learning from Images . International Conference on Machine Learning (ICML)
Adhikary, S. and Boots, B., (2022). Modular Policy Composition with Policy Centroids. Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM)
Received RLDM Travel Award (Declined due to COVID-19)
Adhikary, S. and Boots, B., (2022). Modular Policy Composition with Policy Centroids. Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM)
Received RLDM Travel Award (Declined due to COVID-19)
2021-2022
Geometry-Aware Sampling using Kernel Herding
Extended the kernel herding algorithm to the task of drawing samples from probability distributions over data-spaces corresponding to various structured Riemannian manifolds routinely encountered in robotics.
Papers
Extended the kernel herding algorithm to the task of drawing samples from probability distributions over data-spaces corresponding to various structured Riemannian manifolds routinely encountered in robotics.
Papers
Adhikary, S. and Boots, B., (2022). Sampling over Riemannian Manifolds with Kernel Herding. IEEE International Conference on Robotics and Automation (ICRA)
Awarded Best Paper at R:SS Workshop on Geometry and Topology in Robotics !
Awarded Best Paper at R:SS Workshop on Geometry and Topology in Robotics !
2019-Present
Quantum-Inspired Probabilistic Modeling
Established formal equivalencies between Hidden Quantum Markov Models (HQMMs), a quantum-inspired probabilistic model for sequential data, and well-known models in classical stochastic processes, weighted automata, and uniform quantum tensor networks. Additionally, developed a new approach to learning HQMMs that exploits its parameterization on the Stiefel manifold.
Papers
Established formal equivalencies between Hidden Quantum Markov Models (HQMMs), a quantum-inspired probabilistic model for sequential data, and well-known models in classical stochastic processes, weighted automata, and uniform quantum tensor networks. Additionally, developed a new approach to learning HQMMs that exploits its parameterization on the Stiefel manifold.
Papers
Srinivasan S., Adhikary S., Miller, J., Pokharel, B., Rabusseau, G. and Boots, B., (2021) Towards a Trace-Preserving Tensor Network Representation of Quantum Channels. Second Workshop on Quantum Tensor Networks in Machine Learning, NeurIPS
Adhikary S.∗, Srinivasan S.∗, Miller J., Rabusseau G., & Boots B. (2021) Quantum Tensor Networks, Stochastic Processes, & Weighted Automata. International Conference on Artificial Intelligence and Statistics (AISTATS).
Adhikary, S.∗, Srinivasan, S.∗, Gordon, G. & Boots, B. (2020) Expressiveness and Learning of Hidden Quantum Markov Models. International Conference on Artificial Intelligence and Statistics (AISTATS).
Adhikary S.∗, Srinivasan S.∗, Miller J., Rabusseau G., & Boots B. (2021) Quantum Tensor Networks, Stochastic Processes, & Weighted Automata. International Conference on Artificial Intelligence and Statistics (AISTATS).
Adhikary, S.∗, Srinivasan, S.∗, Gordon, G. & Boots, B. (2020) Expressiveness and Learning of Hidden Quantum Markov Models. International Conference on Artificial Intelligence and Statistics (AISTATS).
2017-2019
Predicting Post-transplant Outcomes in Renal Transplant Patients
Collaborated with clinical experts to develop machine learning models predicting transplant failures, readmissions, and mortality in renal transplant patients.
Papers
Collaborated with clinical experts to develop machine learning models predicting transplant failures, readmissions, and mortality in renal transplant patients.
Papers
Hogan, J., Arenson, M. D., Adhikary, S., Li, K., Zhang, X., Zhang, R., Valdez, J. N., Lynch, R. J., Sun, J., Adams, A. B., & Patzer, R. E. (2019). Assessing Predictors of Early and Late Hospital Readmission After Kidney Transplantation. Transplantation Direct 5(8)
Teaching
Oct-Dec 2020
Teaching Assistant, University of Washington
CSE 599: Reinforcement Learning
CSE 599: Reinforcement Learning
Dec 2018-May 2019
Teaching Assistant, Georgia Tech.
CS4002: Robots and Society
CS4002: Robots and Society
Aug-Dec 2017
Teaching Assistant, Georgia Tech.
CS4001: Computing, Society, and Ethics
CS4001: Computing, Society, and Ethics