Suvansh Sanjeev
AI Researcher and Engineer working on advanced AI applications
Summary
Work
Education
Projects
Writing
Learning Parameter-Efficient Markovian Quadrotor Dynamics Models
December 1, 2022Master's Thesis focusing on addressing the challenges of quadrotor operation through high-speed maneuvers by developing a two-stage approach to fitting residual dynamics and retaining a concise Markovian state.
Scalable Learning of Safety Guarantees for Autonomous Systems using Hamilton-Jacobi Reachability
January 1, 2021Research presented at Robotics: Science and Systems (RSS) 2021, co-authored, focusing on scalable learning of safety guarantees for autonomous systems.
Ecological Reinforcement Learning
January 1, 2019Research presented at the Deep RL Workshop at NeurIPS 2019, co-authored, exploring ecological reinforcement learning.
Guiding Policies with Language via Meta-Learning
January 1, 2019Research presented at the International Conference on Learning Representations (ICLR) 2019, co-authored, and awarded Best Paper at the Meta-Learning Workshop at NeurIPS 2018.
Embedding Adapters
A technical report detailing research on rigorously evaluating the effectiveness of applying linear transforms to embeddings to improve retrieval performance in practical AI applications.