Mehrdad Zakershahrak
Papers
8
Total Citations
73
H-Index
5
About
Mehrdad Zakershahrak is an AI researcher specializing in explainable artificial intelligence (XAI) and human-robot teaming, with a particular focus on how robotic agents can communicate their decision-making processes effectively to human collaborators. His work sits at a compelling intersection of planning, cognitive science, and robotics, addressing one of the field's most pressing challenges: making AI systems transparent and trustworthy partners for humans. Zakershahrak's most influential contribution, "Interactive Plan Explicability in Human-Robot Teaming" (2018, 23 citations), established foundational frameworks for how robots can model human expectations and generate behaviorally intelligible plans. Building on this, his research on online explanation generation introduced real-time methods for robotic agents to justify their actions as situations unfold, accumulating over 14 citations. A particularly innovative thread in his work examines the ordering and progressive delivery of explanations, recognizing that cognitive load matters as much as correctness — not just *what* a robot explains, but *how* it sequences that information for human comprehension. His application of reinforcement learning to hierarchical explanation generation further demonstrates his commitment to adaptive, human-centered AI. Collectively, his publications signal a researcher dedicated to building AI teammates that are not merely capable, but genuinely comprehensible.
Research Focus
Key Achievements
Top Papers
- 1Interactive Plan Explicability in Human-Robot Teaming23 citations · 2018
- 2Online Explanation Generation for Planning Tasks in Human-Robot Teaming14 citations · 2020
- 3
- 4Online Explanation Generation for Human-Robot Teaming9 citations · 2019
- 5Interactive Plan Explicability in Human-Robot Teaming6 citations · 2019
- 6Progressive Explanation Generation for Human-robot Teaming4 citations · 2019
- 7
- 8