A. Ninomiya
Papers
2
Total Citations
27
H-Index
2
About
A. Ninomiya is a pioneering researcher in adaptive robotics and reinforcement learning, whose work has fundamentally advanced how autonomous agents build and refine their understanding of complex environments. Her primary research areas include state-space construction, neural network-based learning, and robot navigation. Ninomiya’s most significant contribution is the development of an incremental state-space construction method that leverages Adaptive Resonance Theory (ART) neural networks, inspired by Piaget’s concept of contradiction. This innovative approach allows agents to dynamically create and adjust their internal state representations as they encounter new or conflicting information, rather than relying on a fixed, pre-defined state space. Her landmark 2002 paper, “Adaptive state construction for reinforcement learning and its application to robot navigation problems” (17 citations), demonstrated how this method enables robots to navigate effectively by mapping sensory inputs to meaningful states and resolving contradictions. Her related work, “An Incremental State-Space Construction Based on the Notion of Contradiction for Reinforcement Learning” (10 citations), further formalized this framework. By solving the critical challenge of state representation in reinforcement learning, Ninomiya has provided a foundational tool for creating more flexible, intelligent, and autonomous robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2