Abstract: Collision avoidance decision-making (CADM) has a significant potential for marine robotics across diverse applications. However, the existing methods based on reinforcement learning fail to ...
Abstract: Satellite-terrestrial integrated networks (STINs) require a robust handover mechanism to ensure reliable mobility management and load balancing. However, many studies still focus on ...
Roblox scripting blends creativity, optimization, and security to create engaging, stable experiences. By mastering Luau, applying smart performance tweaks, and enforcing server-side logic, developers ...
This project investigates the performance of Proximal Policy Optimization (PPO) and seven algorithmic modifications against multiple reinforcement learning baselines. All experiments are conducted in ...
ABSTRACT: This study presents a modified primal-dual interior point method (MPD-IPM) for solving convex quadratic optimization problems. The modification is performed through linearization of the ...
This repository provides the implementation of ViPO (Visual Preference Policy Optimization) for visual generation. Recent GRPO-based visual alignment pipelines usually optimize a single scalar reward ...