View publication

Self-play has powered breakthroughs in two-player and multi-player games. Here we show that self-play is a surprisingly effective strategy in another domain. We show that robust and naturalistic driving emerges entirely from self-play in simulation at unprecedented scale -- 1.6~billion~km of driving. This is enabled by GigaFlow, a batched simulator that can synthesize and train on 42 years of subjective driving experience per hour on a single 8-GPU node. The resulting policy achieves state-of-the-art performance on three independent autonomous driving benchmarks. The policy outperforms the prior state of the art when tested on recorded real-world scenarios, amidst human drivers, without ever seeing human data during training. The policy is realistic when assessed against human references and achieves unprecedented robustness, averaging 17.5 years of continuous driving between incidents in simulation.

Related readings and updates.

The Entity-Deduction Arena: A Playground for Probing the Conversational Reasoning and Planning Capabilities of LLMs

LLMs are currently effective at answering questions that are clearly asked. However, they may encounter difficulties when faced with ambiguous queries. This emphasizes the need for the development of intelligent agents capable of asking clarification questions, which require complex understanding, state tracking, and planning in multi-turn conversations. In this paper, we study a surrogate problem by employing entity-deducing games as evaluation…
See paper details

Towards Learning Multi-Agent Negotiations via Self-Play

Making sophisticated, robust, and safe sequential decisions is at the heart of intelligent systems. This is especially critical for planning in complex multi-agent environments, where agents need to anticipate other agents' intentions and possible future actions. Traditional methods formulate the problem as a Markov Decision Process, but the solutions often rely on various assumptions and become brittle when presented with corner cases. In…
See paper details