Testing in Real: Sample-efficient failure discovery of contextual failures with Bayesian active learning
Discovering failures that cannot be expressed analytically, in limited samples, using coverage-driven active learning
Posts by Collection
portfolio
Testing in sim: Failure discovery with sampling-based techniques
Improving the efficiency of sampling based evaluation using data-driven techniques
Testing in Sim and Real: Failure prediction using both sources
Combining sample-extensive failure search in sim and sample efficient exploration in real-world platform for model validation
publications
Actuator fault-tolerant control study of an underwater robot with four rotatable thrusters
Journal of Ocean Engineering, 2020
Design and analysis of a fault tolerant control for an AUtonomous Underwater Vehicle (AUV) with four rotatable thrusters
Recommended citation: J. Kadiyam, A. Parashar, D. Deskhukh, S. Mohan (2020). "Actuator fault-tolerant control study of an underwater robot with four rotatable thrusters." Journal of Ocean Engineering, Science Direct, 2020.
Download Paper
Accelerated Algorithms for a Class of Optimization Problems with Constraints
Conference of Decision and Control (CDC), 2022
We investigate the application of higher-order optimization techniques for convex optimization with constraints
Recommended citation: A. Parashar, P. Srivastava, A.M. Annaswamy (2022). "Accelerated Algorithms for a Class of Optimization Problems with Constraints." Conference of Decision and Control (CDC) 2022.
Download Paper
Accelerated Algorithms for a Class of Optimization Problems with Equality and Box Constraints
American Control Conference (ACC), 2023
We investigate the application of higher-order optimization techniques for a class of constrained optimization problems with linear constraints
Recommended citation: A. Parashar, P. Srivastava, A.M. Annaswamy (2023). "Accelerated Algorithms for a Class of Optimization Problems with Equality and Box Constraints." American Control Conference 2023.
Download Paper
Failure Prediction from Limited Hardware Demonstrations
Bayesian Decision making and Uncertainity workshop (BDU), NeurIPS 2024, Allerton, 2024
Novel failure discovery method that combines information from simulation and real-world systems using sampling based tchniques and Bayesian Experimental Design
Recommended citation: A. Parashar, K. Garg, C. Fan (2024). "Failure Prediction from Limited Hardware Demonstrations." IEEE proceedings, Allerton 2025.
Download Paper
On the stability of gradient descent with second order dynamics for time-varying cost functions
Transactions on Machine Learning Research, 2025
Provable stability for second-order gradient based optimization for time-varying cost functions
Recommended citation: T.E. Gibson, S. Acharya, A. Parashar, J.E. Gaudio, A.M. Annaswamy (2025). "On the stability of gradient descent with second order dynamics for time-varying cost functions; Transactions on Machine Learning Research.
Download Paper
RADIUM: Predicting and Repairing End-to-End Robot Failures using Gradient-Accelerated Sampling
IEEE Transactions on Robotics (TRO), 2025
Gradient-accelerated sampling for failure discovery and repair using first order Langevin algorithm
Recommended citation: C. Dawson, A. Parashar, C. Fan (2025). "RADIUM: Predicting and Repairing End-to-End Robot Failures using Gradient-Accelerated Sampling." IEEE Transactions on Robotics 2025.
Download Paper
Learning-based Bayesian Inference for Testing of Autonomous Systems
International Conference on Robotics and Automation (ICRA), 2025
We use learning based techniques to improver the efficiency of MCMC sampling for testing of autonomous systems
Recommended citation: A. Parashar, J. Yin, C. Dawson, P. Tsiotras, C. Fan (2024). "Learning-based Bayesian Inference for Testing of Autonomous Systems." IEEE Robotics and Automation Letters.
Download Paper
Cost-aware Discovery of Contextual Failures using Bayesian Active Learning
Conference on Robot Learning (CoRL), 2025
This work uses coverage-based active learning for discovering failures in robotic systems with a limited sample budget.
Recommended citation: A. Parashar, J. Zhang, Y. Li, C. Fan (2025). "Cost-aware Discovery of Contextual Failures using Bayesian Active Learning." 9th Conference on Robot Learning.
Download Paper
talks
Talk 1 on Relevant Topic in Your Field
Published:
This is a description of your talk, which is a markdown file that can be all markdown-ified like any other post. Yay markdown!
Conference Proceeding talk 3 on Relevant Topic in Your Field
Published:
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.