Search papers, labs, and topics across Lattice.
3
0
4
0
Forget black-box policies: CSRO uses LLMs to generate human-readable code policies in multi-agent RL, achieving performance competitive with traditional methods.
The Random-Offerer mechanism, a staple in bilateral trade, can be even *more* inefficient than we thought – AI-guided search just found a distribution where it captures 2.0749x less gains from trade than the optimal mechanism.
LLMs can autonomously discover novel MARL algorithms that outperform hand-designed baselines, revealing untapped potential in automated algorithm design.