Sungsoo Ahn

I am a postdoctoral research associate at Mohamed bin Zaeyed University of Artificial Intelligence (MBZUAI), supervised by Professor Le Song and Professor Eric Xing. Before this, I received a PhD at Korea Advanced Institute of Science and Technology (KAIST), where I was supevised by Professor Jinwoo Shin. My research mainly focuses on in designing algorithms for graph-structured machine learning problems. See my google scholar page and CV for more information.

Contact: peter [dot] ahn [at] mbzuai [dot] ac [dot] ae

Publications (C: Conference / J: Journal / W: Workshop / P: Preprint)

[C12] Self-Improved Retrosynthetic Planning

  • Junsu Kim, Sungsoo Ahn, Hankook Lee, and Jinwoo Shin

  • International Conference of Machine Learning (ICML) 2021

[C11/W2] RetCL: A Selection-based Approach for Retrosynthesis via Contrastive Learning

  • Hankook Lee, Sungsoo Ahn, Seung-Woo Seo, You Young Song, Eunho Yang, Sung Ju Hwang, and Jinwoo Shin

  • International Joint Conference of Artificial Intelligence (IJCAI 2021)

[C10] A Deeper Look at the Layerwise Sparsity of Magnitude-based Pruning [pdf]

  • Jaeho Lee, Sejun Park, Sangwoo Mo, Sungsoo Ahn, Jinwoo Shin

  • International Conference on Learning Representations (ICLR) 2021

[P1] QOPT: Optimistic Value Function Decentralization
for Cooperative Multi-Agent Reinforcement Learning [pdf]

  • Kyunghwan Son, Sungsoo Ahn, Roben Delos Reyes, Jinwoo Shin, and Yung Yi

  • Arxiv 2020

[C9] Learning from Failure: Training Debiased Classifier from Biased Classifier [pdf, code]

  • Junhyun Nam, Hyuntak Cha, Sungsoo Ahn, Jaeho Lee, and Jinwoo Shin

  • Neural Information Processing Systems (NeurIPS) 2020

[C8] Guiding Deep Molecular Optimization with Genetic Exploration [pdf, code]

  • Sungsoo Ahn, Junsu Kim, Hankook Lee, Jinwoo Shin

  • Neural Information Processing Systems (NeurIPS) 2020

[C7] Learning What to Defer for Maximum Independent Sets [pdf, code]

  • Sungsoo Ahn, Younggyo Seo, and Jinwoo Shin

  • International Conference of Machine Learning (ICML) 2020

[C6/W1] Variational Information Distillation for Knowledge Transfer [pdf, code]

  • Sungsoo Ahn, Shell Hu, Andreas Damianou, Neil Lawrence, and Zhenwen Dai

  • IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR) 2019

  • Preliminary version appeared at Neural Information Processing Systems (NeurIPS) 2018, workshop on continual learning

[C5] Bucket-Renormalization for Approximate Inference [pdf, code]

  • Sungsoo Ahn, Michael Chertkov, Adrian Weller, and Jinwoo Shin

  • International Conference of Machine Learning (ICML) 2018

  • Journal of Statistical Mechanics: Theory and Experiment (JSTAT) 2019, Machine Learning Special Issue

[J1] Maximum Weight Matching using Odd-sized Cycles: Max-Product Belief Propagation and Half-Integrality [pdf]

  • Sungsoo Ahn*, Sejun Park*, Michael Chertkov*, Andrew E. Gelfand*, and Jinwoo Shin* (*=alphabetical order)

  • IEEE Transactions on Information Theory, 2018

[C4] Gauged Mini-Bucket Elimination for Approximate Inference [pdf, code]

  • Sungsoo Ahn, Michael Chertkov, Jinwoo Shin, and Adrian Weller

  • International Conference on Artificial Intelligence and Statistics (AISTATS) 2018

[C3] Gauging Variational Inference [pdf]

  • Sungsoo Ahn, Michael Chertkov, and Jinwoo Shin

  • Neural Information Processing Systems (NeurIPS) 2017 as poster presentation

  • Journal of Statistical Mechanics: Theory and Experiment (JSTAT) 2019, Machine Learning Special Issue

[C2] Synthesis of MCMC and Belief Propagation [pdf]

  • Sungsoo Ahn, Michael Chertkov, and Jinwoo Shin

  • Neural Information Processing Systems (NeurIPS) 2016 as oral presentation

[C1] Minimum Weight Perfect Matching via Blossom Belief Propagation [pdf]

  • Sungsoo Ahn, Sejun Park, Michael Chertkov, and Jinwoo Shin

  • Neural Information Processing Systems (NeurIPS) 2015 as spotlight presentation

Selected Invited Talks

Scaling Deep Reinforcement Learning to Large Combinatorial Problems

  • AI & CSE Seminar at Pohang University of Science and Technology (POSTECH), Pohang, South Korea (Feb. 2020)

Bucket-Renormalization for Approximate Inference

  • Robotics Seminar at University of Oxford, Oxford, England (July. 2018)

Mini Bucket-Renormalization

  • Physics Informed Machine Learning Workshop, Santa Fe, New Mexico, U.S (Feb. 2018)

Gauge transformation of Graphical Models

  • CNLS Student Seminar at Los Alamos National Laboratory (LANL), Los Alamos, New Mexico, U.S (July. 2017)

Optimizing Gauge Trasformation for Inference in Graphical Model

  • Banff Workshop on Optimization and Inference for Physical Flows on Networks, Banff, Alberta, Canada (Feb. 2017)

Synthesis of MCMC and Belief Propagation

  • CNLS Student Seminar at Los Alamos National Laboratory (LANL), Los Alamos, New Mexico, U.S (July. 2016)

Minimum Weight Perfect Matching via Blossom Belief Propagation

  • Discrete Math Seminar at Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea (Nov. 2015)

Academic Activities

Conference reviewer

  • Neural Information Processing Systems (NeurIPS): 2018 / 2019 / 2020

  • International Conference of Machine Learning (ICML): 2019 / 2020 / 2021

  • International Conference of Learning Representations (ICLR): 2019 / 2020

Journal reviewer

  • Journal of Machine Learning Research (JMLR)

  • IEEE/ACM Transactions on Networking

Experience

Amazon Cambridge Development Center

  • Cambridge, England (June. 2018 - Aug. 2018)

  • Research intern (supervised by Zhenwen Dai)

University of Cambridge

  • Cambridge, England (Mar. 2018 - May. 2018)

  • Visiting student (hosted by Adrian Weller)

Los Alamos National Laboratory

  • Los Alamos, New Mexico, U.S (June. 2017 - Aug. 2017)

  • Research intern (mentored by Michael Chertkov)

Los Alamos National Laboratory

  • Los Alamos, New Mexico, U.S (June. 2016 - Aug. 2016)

  • Research intern (mentored by Michael Chertkov)

Penta Security Systems Inc.

Seoul, Korea (Jan. 2014 - Feb. 2014)

Software Developer Intern