Skip to main content

Chuanbo Hua

Postdoctoral Fellow
InnoCore PRISM-AI Project
Korea Advanced Institute of Science & Technology (KAIST)

GitHub: cbhua
Scholar: Chuanbo Hua
Email: cbhua [at] kaist.ac.kr
Phone: (+82) 010-6608-6548
Address: 4217, E2-1, 291 Daehak-ro, Yuseong-gu, Daejeon, South Korea

"Education does for the mind, experience does for the soul."


🗓  Last Update: 2025.10.20.

💡  Research Interests: reinforcement learning, combination optimization, LLM for algorithm design.


I am a postdoctoral fellow at Korea Advanced Institute of Science & Technology (KAIST) . I'm doing research with my colleagues at the Systems Intelligence Laboratory (SILab) suprevised by Prof. Jinkyoo Park. I am the co-founder of the AI4CO open source group and work closely with the DiffEqML open source group.

The primary focus of my ongoing research is the development of reinforcement learning (RL) algorithms for the study of combinatorial optimization (CO) problems. We create an extensive open-source benchmark RL4CO to provide a unified framework for RL-based CO algorithms, and to facilitate reproducible research in this field, decoupling the science from the engineering. I also explore machine learning methods for dynamical physical system simulation, scientific computing, and trajectory prediction. Recent interests include reinforcement learning for continuous dynamical system control, robot navigation and high efficient complex physical system simulation.

Find my works:  Publications  ∣  Research  ∣  Projects
Recent News
10/05/2025.
Paper " PARCO: Learning Parallel Autoregressive Policies for Efficient Multi-agent Combinatorial Optimization" got accepted at the NeurIPS 2025 (Poster).
09/28/2025.
Paper " RRNCO: Neural Combinatorial Optimization for Real-world Routing" got accepted at the NeurIPS 2025 Workshop (Poster).
09/27/2025.
Paper " PILLM: Physics-Informed Large Language Models for HVAC Anomaly Detection with Autonomous Rule Generation" got accepted at the NeurIPS 2025 Workshop (Oral).
09/01/2025.
I started my postdoctoral fellowship at KAIST, under the project of InnoCore PRISM-AI.
06/07/2025.
Paper " Routefinder: Towards Foundation Models for Vehicle Routing Problems" got accepted at the TMLR .
05/15/2025.
Paper " RL4CO: an Extensive Reinforcement Learning for Combinatorial Optimization Benchmark." got accepted at the KDD Dataset & Benchmark Track (Oral).
04/02/2025.
Paper " TrajEvo: Designing Trajectory Prediction Heuristics via LLM-driven Evolution" got accepted at the ICML 2025 Workshop (Poster).
03/28/2025.
Paper " BuildEvo: Designing Building Energy Consumption Forecasting Heuristics via LLM-driven Evolution" got accepted at the ICML 2025 Workshop (Poster).
01/24/2025.
Paper " CAMP: Collaborative Attention Model with Profiles for Vehicle Routing Problems " got accepted at the AAMAS 2025 (Oral).
01/05/2025.
Paper " Accelerating Chiplet Placement & Routing Optimization with Machine Learning " got accepted at the DesignCon 2025 (Best Paper).
12/10/2024.
Paper "AoP-SAM: Automation of Prompts for Efficient Segmentation" got accepted at the AAAI 2025 (Poster).
09/24/2024.
Paper "Deep learning-based coagulant dosage prediction for extreme events leveraging large-scale data" got accepted at the JWPE (IF: 7).
07/15/2024.
Paper "RouteFinder: Towards Foundation Models for Vehicle Routing Problems" got accepted at the ICML 2024 Workshop (Oral).
06/01/2024.
Paper "Deep learning-based coagulant dosage prediction for extreme events leveraging large-scale data" got accepted at the JWPE (IF: 7).
02/13/2024.
Paper "HiMAP: Learning Heuristics-Informed Policies for Large-Scale Multi-Agent Pathfinding" got accepted at the AAMAS 2024 (Poster).
02/01/2024.
Paper "Optimizing coagulant dosage using deep learning models with large-scale data" got accepted at the Chemosphere (IF: 8.1).
10/31/2023.
Paper "RL4CO: a Unified Reinforcement Learning for Combinatorial Optimization Library" got accepted at the NeurIPS 2023 Workshop (Oral).
09/30/2023.
Paper "Learning Efficient Surrogate Dynamic Models with Graph Spline Networks" got accepted at the NeurIPS 2023 (Poster).
08/01/2023.
Paper "Comparing artificial and deep neural network models for prediction of coagulant amount and settled water turbidity" got accepted at the JWPE (IF: 7).
04/01/2023.
Paper "Coagulant dosage determination using deep learning-based graph attention multivariate time series forecasting model" got accepted at the Water Research (IF: 12.8).
09/01/2022.
I started my Ph.D. in KAIST advised by Prof. Jinkyoo Park.
06/15/2022.
Paper "Efficient Continuous Spatio-Temporal Simulation with Graph Spline Networks" got accepted at the ICML 2022 Workshop (Oral).

Mailing Address
Industrial & Systems Engineering, Korea Advanced Institute of Science & Technology (KAIST)
4217, E2-1, 291 Daehak-ro, Yuseong-gu, Daejeon, South Korea