20.500.12556/RUL-181718 Scenario-guided temporal prototypes in reinforcement learning Deep reinforcement learning policies are hard to deploy in safety-critical settings, because they fail to explain why a sequence of actions is taken. We introduce an intrinsically interpretable framework that learns compact summaries of recurring behavior and uses them for case-based decision making. Our method (i) discovers global regimes by grouping trajectories into a small set of recurrent patterns and (ii) learns a prototype-conditioned local policy that maps the current short-horizon pattern to an action (“this matches prototype X → take action Y”). Each action is accompanied by a similarity score to relevant prototypes, which provide the explanations. We evaluate our approach on two domains: (1) CarRacing (pixel-based continuous control) and (2) a real voltage-control problem in low-voltage distribution networks. Our results indicate that the method provides clear pre hoc explanations while keeping task performance close to the reference policy. deep reinforcement learning explainability prototypes globoko spodbujevano učenje razložljivost prototipi true false true Angleški jezik Slovenski jezik Članek v reviji 2026-04-14 12:23:38 2026-04-14 12:23:39 2026-04-15 04:25:17 0000-00-00 00:00:00 2026 0 0 14 str. iss. 1, art. 21 Vol. 8 Jan. 2026 0000-00-00 Zaloznikova Objavljeno apc 0000-00-00 0000-00-00 0000-00-00 004.8:007.52 2504-4990 10.3390/make8010021 265133315 RAZ_Dobravec_Blaz_2026.pdf RAZ_Dobravec_Blaz_2026.pdf 1 6DC0A53A40F17F8241D59BED35D35578 3ebfd2eb8b9a6fe40311cc86674109ccfcd9c1744b46dbf89f1d36ebfb37196a 137e5f99-37ec-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=232409 https://www.mdpi.com/2504-4990/8/1/21 1 c89c4e5f-37eb-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=232408 Fakulteta za računalništvo in informatiko 0 0 0