Decentralized Data Collection for Robotic Fleet Learning: A Game-Theoretic Approach
Decentralized Data Collection for Robotic Fleet Learning: A Game-Theoretic Approach
复制标题
DOI：
--
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Oguzhan Akcin;Po-han Li;Shubhankar Agarwal;Sandeep P. Chinchali
中科院分区：
文献类型：
--
作者：
Oguzhan Akcin;Po-han Li;Shubhankar Agarwal;Sandeep P. Chinchali
联网的自动驾驶汽车（AV）收集TB级的传感数据，这些数据通常被传输到中央服务器（“云”），用于训练机器学习（ML）模型。理想情况下，这些机器人应该上传所有数据，特别是来自罕见操作环境的数据，以便训练强大的ML模型。然而，由于高昂的网络带宽和数据标签成本，这是不可行的。相反，我们提出了一种合作数据采样策略，在这种策略中，地理分布的AV合作在云中收集不同的ML训练数据集。由于AV具有共享的目标，但关于彼此的本地数据分布和感知模型的信息最少，因此我们可以自然地将合作数据收集视为N人数学游戏。我们表明，我们的合作采样策略使用最少的信息收敛到一个集中的预言机的政策，所有AV的完整信息。此外，我们从理论上描述了与贪婪采样相比，我们的博弈论策略的性能贝内。最后，我们通过实验证明，我们的方法优于标准基准高达21。在4个感知数据集上的9%，包括在恶劣天气条件下的自动驾驶。至关重要的是，我们在真实世界数据集上的实验结果与我们的理论保证密切一致。
: Fleets of networked autonomous vehicles (AVs) collect terabytes of sensory data, which is often transmitted to central servers (the “cloud”) for training machine learning (ML) models. Ideally, these fleets should upload all their data, especially from rare operating contexts, in order to train robust ML models. However, this is infeasible due to prohibitive network bandwidth and data labeling costs. Instead, we propose a cooperative data sampling strategy where geo-distributed AVs collaborate to collect a diverse ML training dataset in the cloud. Since the AVs have a shared objective but minimal information about each other’s local data distribution and perception model, we can naturally cast cooperative data collection as an N -player mathematical game. We show that our cooperative sampling strategy uses minimal information to converge to a centralized oracle policy with complete information about all AVs. Moreover, we theoretically characterize the performance benefits of our game-theoretic strategy compared to greedy sampling. Finally, we experimentally demonstrate that our method outperforms standard benchmarks by up to 21 . 9% on 4 perception datasets, including for autonomous driving in adverse weather conditions. Crucially, our experimental results on real-world datasets closely align with our theoretical guarantees.