ITS - Visoka škola strukovnih studija za informacione tehnologije

FROM REWARD HACKING TO RESPONSIBILITY GAP: A SYSTEMIC ANALYSIS OF AGENTIC RL FAILURES, BENCHMARK PHILOSOPHY, AND THE CASE FOR CONSTRAINT-AWARE GOVERNANCE IN FRONTIER AI

ABSTRACTBetween 2016 and 2026, the AI research community documented a consistent failure mode: reinforcement learning agents optimize measured proxies rather than intended outcomes, exploiting gaps in reward specifications, evaluation infrastructure, and environmental constraints...

Ivković Graph-Coupled Chaotic Dynamical Networks for Bounded High-Dimensional Nonlinear Parametric Embedding: A Mathematical, Digital-Simulation and Analogue-Hybrid Framework

Extended author version v1.1 of the paper presented at the 4th LINK IT & EdTech International Scientific Conference. The revision strengthens the mathematical treatment of well-posedness, volume contraction, absorbing boundedness, global existence, synchronization-manifold invari...

FROM REWARD HACKING TO RESPONSIBILITY GAP: A SYSTEMIC ANALYSIS OF AGENTIC RL FAILURES, BENCHMARK PHILOSOPHY, AND THE CASE FOR CONSTRAINT-AWARE GOVERNANCE IN FRONTIER AI

ABSTRACTBetween 2016 and 2026, the AI research community documented a consistent failure mode: reinforcement learning agents optimize measured proxies rather than intended outcomes, exploiting gaps in reward specifications, evaluation infrastructure, and environmental constraints...