publications

2025

1. RLC 2025

   Joint-Local Grounded Action Transformation for Sim-to-Real Transfer in Multi-Agent Traffic Control

   Justin Turnau, Longchao Da, Khoa Vo, and 4 more authors

   Reinforcement Learning Journal, 2025

   Abs HTML

   Addresses the sim-to-real gap in multi-agent reinforcement learning for traffic signal control, proposing a joint-local grounded action transformation that enables policies trained in simulation to transfer effectively to real-world scenarios. Presented at the Reinforcement Learning Conference (RLC) 2025.

2. arXiv

   

   Generative AI in Transportation Planning: A Survey

   Longchao Da, Tiejin Chen, Zhuoheng Li, and 14 more authors

   arXiv preprint arXiv:2503.07158, 2025

   Abs HTML

   A comprehensive survey examining how large language models and generative AI can optimize transportation systems, with a focus on origin-destination matrix estimation and urban planning applications. Under journal revision.

3. Poster

   Compression of Deep Neural Networks for Edge Devices

   Shreyas Bachiraju

   2025

   Abs HTML

   Mixed-precision quantization and model compression for deploying depth-estimation models on NVIDIA Jetson Nano hardware, achieving 33% lower latency, 42% smaller model size, and 50% higher throughput while retaining 99% of baseline accuracy.