只需取整：量化观测空间实现动态运动的内存高效学习
Just Round: Quantized Observation Spaces Enable Memory Efficient Learning of Dynamic Locomotion
AI总结：
本文提出通过量化 DRL 智能体的观测空间降低边缘机器人运动学习的内存开销，并在四个仿真运动任务上用 PPO 与 SAC 验证，可在不损害性能的情况下将内存成本最高降低 4.2 倍。
摘要：
深度强化学习（DRL）是合成复杂机器人行为的最强大工具之一。但训练 DRL 模型在计算和内存方面都极其密集，需要大型训练数据集和经验回放缓冲区才能获得高性能结果。这给下一代需要在边缘端学习以适应环境的野外机器人带来了挑战。本文中，我们通过观测空间量化来着手解决这一问题。我们使用四个仿真机器人运动任务以及两种最先进的 DRL 算法——同策略的近端策略优化（PPO）和异策略的软演员-评论家（SAC）——评估该方法，发现观测空间量化可将总体内存成本降低多达 4.2 倍，且不会影响学习性能。
英文摘要：
Deep reinforcement learning (DRL) is one of the most powerful tools for synthesizing complex robotic behaviors. But training DRL models is incredibly compute and memory intensive, requiring large training datasets and replay buffers to achieve performant results. This poses a challenge for the next generation of field robots that will need to learn on the edge to adapt to their environment. In this paper, we begin to address this issue through observation space quantization. We evaluate our approach using four simulated robot locomotion tasks and two state-of-the-art DRL algorithms, the on-policy Proximal Policy Optimization (PPO) and off-policy Soft Actor-Critic (SAC) and find that observation space quantization reduces overall memory costs by as much as 4.2x without impacting learning performance.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
