按年被提到
20182019 · 2 次2026
强化学习与智能体›深度强化学习与世界模型·arXiv:1810.06544 · cs.LG

Deep Imitative Models for Flexible Inference, Planning, and Control

Nicholas Rhinehart, Rowan McAllister, Sergey Levine · 2018
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2019-06-20 · 最近 2019-10-03
2不同的人
2活跃月份
2讨论串
110被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Imitation Learning (IL) is an appealing approach to learn desirable autonomous behavior. However, directing IL to achieve arbitrary goals is difficult. In contrast, planning-based algorithms use dynamics models and reward functions to achieve goals. Yet, reward functions that evoke desirable behavior are often difficult to specify. In this paper, we propose Imitative Models to combine the benefits of IL and goal-directed planning. Imitative Models are probabilistic predictive models of desirable behavior able to plan interpretable expert-like trajectories to achieve specified goals. We derive families of flexible goal objectives, including constrained goal regions, unconstrained goal sets, and energy-based goals. We show that our method can use these objectives to successfully direct behavior. Our method substantially outperforms six IL approaches and a planning-based approach in a dynamic simulated autonomous driving task, and is efficiently learned from expert demonstrations without online data collection. We also show our approach is robust to poorly specified goals, such as goals on the wrong side of the road.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论cr4zy · 2019-06

“Related recent research that handles decoy waypoints [1][2]. They use imitation learning for path planning directly from sensor data, instead of just using machine learning for perception, tracking, prediction and doing MPC/LQR for the planner. This allows inferring weird paths, perhaps, like "run into this narrow pit stop":) [1] https://arxiv.org/abs/1…”

Tesla Model 3 spoofed off the highway原评论 ↗
被提交为 story · 1 次