2 papers
cs.LG2026
Learning Suffers More Than the Policy Class Under Partial Observability: A Closed-Form Analysis
Idil Gözel
When a reinforcement learning agent cannot observe the full state, we usually blame its policies: it cannot see enough to represent a good one. We show that in a solvable case the…
physics.optics2025
Harnessing the Diamond-Air Interface as an Efficient Photon Antenna for Solid-State Emitters
Paul Weinbrenner, Aina Lopez Benet, İdil Gözel +1
Extracting photons from defect centers is challenging due to the high refractive index of typical substrates. For nitrogen-vacancy centers in diamond, reaching saturation count rat…