Tal Feiner

Elbit Systems (Israel), Ben-Gurion University of the Negev

Papers

Total Citations

H-Index

About

Tal Feiner is a researcher advancing the frontier of autonomous robotics, with a focus on intelligent navigation and semantic perception. Their work bridges the gap between classical robotic algorithms and modern machine learning, particularly through parameter tuning and object-level scene understanding. In their highly cited 2023 paper, "PTDRL: Parameter Tuning Using Deep Reinforcement Learning" (4 citations), Feiner introduced a novel strategy that enables robots to autonomously adapt navigation parameters to new environments, eliminating the need for manual re-tuning and significantly improving operational robustness. Earlier, in their 2020 work "Representing and updating objects' identities in semantic SLAM" (3 citations), Feiner tackled a fundamental challenge in semantic simultaneous localization and mapping by proposing a probabilistic representation of object identities. This approach allows robots to maintain and update beliefs about objects in dynamic environments, enhancing long-term autonomy and scene comprehension. Feiner’s contributions are particularly impactful for researchers working on adaptive robotics and semantic mapping, offering practical frameworks that move beyond static algorithms. Their work exemplifies how integrating deep reinforcement learning and probabilistic reasoning can create more resilient, context-aware robotic systems.