Akshay Narayan
Papers
1
Total Citations
20
H-Index
1
About
Akshay Narayan is a researcher advancing the frontier of artificial intelligence through innovations in hierarchical reinforcement learning (HRL). His work focuses on developing model-based approaches that enable agents to efficiently solve large, complex problems by exploiting shared knowledge and selective execution across multiple levels of abstraction. Narayan’s most-cited paper, “An Efficient Approach to Model-Based Hierarchical Reinforcement Learning” (2017), introduces a novel transition dynamics learning algorithm that identifies common structures within tasks, dramatically reducing sample complexity. This contribution has garnered 20 citations, reflecting its significance in the HRL community. By bridging model-based planning with hierarchical decomposition, Narayan’s research addresses a critical bottleneck in scaling reinforcement learning to real-world applications, such as robotics and autonomous systems. His work stands out for its practical emphasis on efficiency and generalization, offering a pathway to more sample-efficient and interpretable AI. For students and researchers exploring reinforcement learning, Narayan’s contributions provide a foundational framework for tackling high-dimensional decision-making problems, making him a notable voice in the ongoing effort to build more capable and resource-conscious intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1An Efficient Approach to Model-Based Hierarchical Reinforcement Learning20 citations · 2017