Yiren Lu

We will be presenting our paper “Imitation Is Not Enough: Robustifying Imitation with Reinforcement Learning for Challenging Driving Scenarios” at #IROS 2023: arxiv.org/abs/2212.11419. We show imitation and RL complement each other for solving long-tail driving scenarios.

@Waymo @eccvconf

If you attend ECCV in Milan 🇮🇹, come say hi and check out our poster! 10/1 Tue, 4:30-6:30pm CEST

New paper at #ECCV 24 🇮🇹! We present an on-policy multi-turn RL approach with a simple reward function can effectively align pre-trained Transformer-based imitative behavior models with human preferences for safety and reliability. arxiv.org/pdf/2409.18343

Waymo is now open to everyone in SF!

#CVPR2024 Workshop on AV Simulation! Simulation is a crucial tool to accelerate the development of safe autonomous driving (AD). In our full-day CVPR workshop with exciting speakers from Waymo, Tesla, Waabi, Wayve, Nuro, Princeton and many others, we’re bringing together 1/N

Excited about our most recent release of Waymax, a lightweight, hardware-accelerated, multi-agent JAX simulator for autonomous driving planning and sim-agents research!

This is happening now! Join here: waabi-ai.zoom.us/j/81205991343?…

Visualizations available on our webpage: waymo.com/intl/zh-cn/res…

We will be presenting our paper “Imitation Is Not Enough: Robustifying Imitation with Reinforcement Learning for Challenging Driving Scenarios” at #IROS 2023: arxiv.org/abs/2212.11419. We show imitation and RL complement each other for solving long-tail driving scenarios.

Check out our workshop agents4ad.github.io if you are attending #IROS 2023!

I'm excited to publish this in-depth look at the safety of driverless taxis in San Francisco and Phoenix. These vehicles' mistakes have gotten a lot of media attention, but you have to remember that they happened over about 8 million miles of driving. understandingai.org/p/driverless-c…

A milestone for @Waymo and autonomous driving industry!

@ying11231 Congrats!

Thrilled to be selected as an ML and systems rising star! Looking forward to meet everyone at the workshop! Let’s talk about ML efficiency and ML for reasoning, program, and math.

"Waymo is able to drive fully autonomously – with nobody behind the wheel – across all of San Francisco, 24/7" blog.waymo.com/2022/12/wheels…

Waymo: Taking our next step in the City by the Bay blog.waymo.com/2022/03/taking…

Waymo to start charging money for its robotaxi rides in San Francisco theverge.com/2022/3/1/22956… via @Verge