I am a PhD student in Informatics at the University of Edinburgh, supervised by Esmeralda S. Whitammer (also called Nikolay Malkin in older records) and Henry Gouk. My PhD studies are fully funded by ARIA and CIFAR, with additional support from Global Korea Scholarship.
I’m broadly interested in generative models, reinforcement learning (especially, MaxEnt RL), Monte Carlo methods, and their applications in sampling (e.g., Boltzmann generators, sampling from Bayesian posteriors of pretrained generative models), optimisation (e.g., drug discovery, combinatorial optimisation), causality (e.g., causal discovery, causal representation learning), and language modelling (e.g., reasoning, prior elicitation from LLMs, neuro-symbolic methods). Most recently, my research focuses on 1) solving sampling or related problems with amortised sampling and/or Monte Carlo methods, 2) extracting structured knowledge from language models. My long-term goal is understanding and improving deep generative models through a probabilistic and symbolic lens, thereby bringing human-like (hierarchical, compositional, uncertainty-aware, and safe) reasoning capabilities to AI systems.
Here is my CV (last updated in May 2026).
📝 Publications
* indicates equal contribution.
- Structured Inference with Large Language Gibbs [TBU]
Sanghyeok Choi, Henry Gouk, Esmeralda S. Whitammer
ICML 2026 Workshop on Structured Probabilistic Inference & Generative Modeling (SPIGM) - Generalised Latent Slice Sampling [TBU]
Kirill Tamogashev, Sanghyeok Choi, Arran Carter, Víctor Elvira, alix, Esmeralda S. Whitammer
ICML 2026 Workshop on Structured Probabilistic Inference & Generative Modeling (SPIGM) - Amortised Inference through One-Step Implicit Sampling [TBU]
Vincent Pauline, Kirill Tamogashev, Arran Carter, Sanghyeok Choi, Stefan Bauer, Esmeralda S. Whitammer
ICML 2026 Workshop on Structured Probabilistic Inference & Generative Modeling (SPIGM) - Solving Integer Linear Programming with Parallel Tempering [TBU]
Kyuil Sim, Sanghyeok Choi*, Jinkyoo Park*
ICML 2026 Workshop on Structured Probabilistic Inference & Generative Modeling (SPIGM) - Aligning Few-Step Generative Model via Amortizing Sample-Based Variational Inference [TBU]
Jaewoo Lee, Hyeongyu Kang, Dohyun Kim, Kyuil Sim, Woocheol Shin, Minsu Kim, Taeyoung Yun, Jeongjae Lee, Sanghyeok Choi, Tabitha Edith Lee, Jong Chul Ye, Jinkyoo Park
ICML 2026 Workshop on Structured Probabilistic Inference & Generative Modeling (SPIGM) - Discrete Diffusion Samplers and Bridges: Off-Policy Algorithms and Applications in Latent Spaces [paper] [code]
Arran Carter*, Sanghyeok Choi*, Kirill Tamogashev*, Víctor Elvira, Nikolay Malkin
ICML 2026 & ICLR 2026 Workshop on Deep Generative Model in Machine Learning: Theory, Principle and Efficacy - Reinforced sequential Monte Carlo for amortised sampling [paper] [code]
Sanghyeok Choi, Sarthak Mittal, Víctor Elvira, Jinkyoo Park, Nikolay Malkin
ICML 2026 (spotlight), an earlier version was presented at ICML 2025 Workshop on Generative AI and Biology - Diffusion Alignment as Variational Expectation-Maximization [paper]
Jaewoo Lee, Minsu Kim, Sanghyeok Choi, Inhyuck Song, Sujin Yun, Hyeongyu Kang, Woocheol Shin, Taeyoung Yun, Kiyoung Om, Jinkyoo Park
ICLR 2026 - torchgfn: A PyTorch GFlowNet library [paper], [code]
Joseph D Viviano*, Omar G. Younis*, Sanghyeok Choi*, Victor Schmidt, Yoshua Bengio, Salem Lahlou
NeurIPS 2025 Workshop on Frontiers in Probabilistic Inference: Sampling Meets Learning - Rl4co: an extensive reinforcement learning for combinatorial optimization benchmark [paper], [code]
Federico Berto*, Chuanbo Hua*, Junyoung Park*, Laurin Luttmann*, Yining Ma, Fanchen Bu, Jiarui Wang, Haoran Ye, Minsu Kim, Sanghyeok Choi, Nayeli Gast Zepeda, André Hottung, Jianan Zhou, Jieyi Bi, Yu Hu, Fei Liu, Hyeonah Kim, Jiwoo Son, Haeyeon Kim, Davide Angioni, Wouter Kool, Zhiguang Cao, Qingfu Zhang, Joungho Kim, Jie Zhang, Kijung Shin, Cathy Wu, Sungsoo Ahn, Guojie Song, Changhyun Kwon, Kevin Tierney, Lin Xie, Jinkyoo Park
KDD 2025 Datasets and Benchmarks Track (Oral) - Neural Genetic Search in Discrete Spaces [paper] [code]
Hyeonah Kim*, Sanghyeok Choi*, Jiwoo Son, Jinkyoo Park, Changhyun Kwon
ICML 2025 & ICLR 2025 Workshop on Deep Generative Model in Machine Learning: Theory, Principle and Efficacy - Improved Off-policy Reinforcement Learning in Biological Sequence Design [paper] [code]
Hyeonah Kim, Minsu Kim, Taeyoung Yun, Sanghyeok Choi, Emmanuel Bengio, Alex Hernández-García, Jinkyoo Park
ICML 2025 & NeurIPS 2024 Workshop on AI for New Drug Modalities - Adaptive teachers for amortized samplers [paper] [code]
Minsu Kim*, Sanghyeok Choi*, Taeyoung Yun, Emmanuel Bengio, Leo Feng, Jarrid Rector-Brooks, Sungsoo Ahn, Jinkyoo Park, Nikolay Malkin, Yoshua Bengio
ICLR 2025 - Ant Colony Sampling with GFlowNets for Combinatorial Optimization [paper], [code]
Minsu Kim*, Sanghyeok Choi*, Hyeonah Kim, Jiwoo Son, Jinkyoo Park, Yoshua Bengio
AISTATS 2025 - Genetic-guided GFlowNets: Advancing in Practical Molecular Optimization Benchmark [paper], [code]
Hyeonah Kim, Minsu Kim, Sanghyeok Choi, Jinkyoo Park
NeurIPS 2024 - Equity-Transformer: Solving NP-Hard Min-Max Routing Problems as Sequential Generation with Equity Context [paper], [code]
Jiwoo Son*, Minsu Kim*, Sanghyeok Choi, Hyeonah Kim, Jinkyoo Park
AAAI 2024 - Multi-agent reinforcement learning based actuator control for EV HVAC systems [paper]
Sungho Joo, Dongmin Lee, Minseop Kim, Taeho Lee, Sanghyeok Choi, Seungju Kim, Jeyeol Lee, Joongjae Kim, Yongsub Lim, Jeonghoon Lee
IEEE Access 2022
📖 Education
- Sep 2025 - Current, PhD student in Informatics, University of Edinburgh, UK.
- Sep 2023 - Aug 2025, MS in Industrial & Systems Engineering, KAIST, South Korea.
- Mar 2017 - Aug 2023, BBA and BS in Industrial Engineering, Seoul National University, South Korea.
💻 Experience
- Oct 2024 - Dec 2025, Research Intern @ Mila, remotely.
- Sep 2022 - Aug 2023, Undergraduate Resercher @ Systems Intelligence Lab, KAIST.
- Jan 2022 - Aug 2022, Machine Learning Research Engineer (Intern) @ MakinaRocks.
- Mar 2021 - Dec 2021, Undergraduate Researcher @ Vision & Learning Lab, Seoul National University.
- Mar 2019 - Jan 2021, Military Service @ R.O.K Airforce.
⭐ Open Source Projects
🎖 Honors and Awards
- Jul 2025, Global Korea Scholarship, National Institute for International Education (NIIED), Republic of Korea.
- Jul 2020, 1st place in Manufacturing Process Optimization with AI (Competition), DACON & LG Science Park.
- May 2020, 1st place in Nowcast Prediction (Competition), DACON & KIMM.
- Feb 2018, Bang Il-Young Scholarship (Full-tuition and living stipend for 3 years), The Bang Il-Young Foundation.
- Aug 2017, Scholarship for Academic Excellence (Full-tuition for 1 semester), Seoul National University.