MIT's Cathy Wu boosts RL training efficiency 30x to design smarter transportation systems
MIT News AI · rss · 2026-10-03
MIT associate professor Cathy Wu discusses her lab's use of reinforcement learning to design transportation and other complex systems.
- Why RL: Designing transportation systems requires analyzing hundreds to thousands of variants—beyond evidence-driven tools; RL promises exponential efficiency gains.
- Setbacks and breakthrough: After a viral 2018 result applying RL to autonomous-vehicle traffic impact, two years of failed attempts followed. In 2023 her team found RL trains poorly on 90% of related problems but well on 10%—training only on those and combining models improved training efficiency up to 30x (100 models may become 3).
- Policy impact: Recent work shows eco-driving (intelligently controlling vehicle speeds) could cut vehicle emissions by 11–22%, demonstrating RL can inform real transportation policy.
- Approach: She champions "use-inspired basic research," letting consequential problems shape research direction.
More from Research
- Symbolic supervision gives video models equal reasoning at ~1/30 the compute, Rubik's Cube study finds — niloofar_mire · 2026-10-03
- Sony AI open-sources PAVAS, a physics-aware video-to-audio model accepted as CVPR 2026 Oral — mittu1204 · 2026-10-03
- Maryland RCT: GPT-4o tutor offering left students 4 points lower, 15% used it — lmoroney · 2026-10-03
- OpenAI paper: simulating deployment predicts LLM safety before release — JacquesThibs · 2026-10-03
- PyTorch Helion DSL integrated into vLLM beats CUTLASS and DeepGEMM with 10%+ throughput gains — PyTorch · 2026-10-03
- Neural memory as an unbounded 2D lattice: writes and reads through a tiny 3x3 bump — CFGeek · 2026-10-03