Takatomo Mihana
Papers
1
Total Citations
3
H-Index
1
About
Takatomo Mihana is a leading researcher in the emerging field of photonic intelligence, where he harnesses the inherent complexity of semiconductor lasers to solve real-world computational problems. His primary contributions lie at the intersection of nonlinear laser dynamics and decision-making theory, most notably in developing novel approaches to the multi-armed bandit problem—a classic challenge in reinforcement learning. In his highly cited 2021 work, Mihana demonstrated how chaotic temporal waveforms from a semiconductor laser can be used for adaptive decision-making in environments with time-varying reward probabilities. By employing a tug-of-war method that compares a threshold against the laser's chaotic output, his team showed that photonic systems can efficiently explore and exploit changing conditions without conventional digital processing. This work has garnered significant attention (3+ citations in a niche field) for its elegant proof-of-principle that physical chaos can serve as a computational resource. Mihana’s research is paving the way for ultra-fast, energy-efficient decision-making systems, with potential applications in autonomous robotics, financial trading, and adaptive communications—establishing him as a key innovator in the nascent field of photonic reinforcement learning.
Research Focus
Key Achievements
Top Papers
- 1