Himani Arora

Papers

Total Citations

H-Index

About

Himani Arora’s research lies at the intersection of robotics, human-computer interaction, and machine learning, with a focus on enabling more intuitive and efficient autonomous systems. Her most cited work, “Multi-task Learning for Continuous Control” (2018, 11 citations), addresses a critical bottleneck in reinforcement learning: how to train robotic agents that can generalize across related tasks without catastrophic forgetting. This contribution is foundational for developing robots that learn quickly and adapt to everyday environments, a prerequisite for practical, real-world deployment. Earlier, Arora explored gesture-based interfaces with “3D Accelerometer based Gesture Device for the Recognition of Digits” (2015, 2 citations), designing a compact handheld device for natural human-robot interaction. This work highlights her versatility in bridging hardware and algorithmic innovation. While her citation counts reflect an emerging career, her research tackles fundamental challenges in multi-task learning and intuitive control—areas poised for significant impact as robotics moves toward general-purpose autonomy. Her work is particularly relevant for students and researchers interested in reinforcement learning, robotic manipulation, and human-robot collaboration.