Publications

Under Review

• Scalable Multi-Agent Path Finding via cause-specific local repair — AAAI 2027.
• Risk-sensitive reinforcement learning with generative value distributions — NeurIPS 2026.
• Additional manuscripts under review at AAAI 2027: generative experience replay for continual off-policy RL; robot controller interface calibration; distributionally robust learning under calibrated shifts.

Note: titles are described rather than quoted while the review process is ongoing.