BU, Fanchen 卜凡辰

😊 Hi, I am Fanchen (/fantʂʰən/)! You can also call me Vezen (/vezən/).

I am a PhD student in the Data Mining Lab at KAIST, under Prof. Shin, Kijung.

My research interests lie in graph mining, graph algorithms, and combinatorial optimization.

Besides research, I like music 🎼, math 📐, thinking 🤔, and exercise 🏋️‍♀️!

過則勿憚改 When you make a mistake, don't be afraid to correct yourself
——《論語·學而》


Contact

✉️ Email: boqvezen97@kaist.ac.kr

👾 Github

💼 LinkedIn

Education

Doctor

2022.03 - | Korea Advanced Institute of Science and Technology (KAIST)

  • Electrical Engineering

Researcher

2021.09 - 2022.02 | Korea Advanced Institute of Science and Technology (KAIST)

Master

2019.09 - 2021.08 | Korea Advanced Institute of Science and Technology (KAIST)

  • Electrical Engineering, GPA: 4.21/4.3
  • Thesis: A Novel Optimization Algorithm with Orthogonality for Deep Neural Networks Inspired by Feedback Integrators

Bachelor

2015.09 - 2019.08 | University of Chinese Academy of Sciences (UCAS)

  • Computer Science and Technology, GPA: 3.71/4

Publications

You can also check them on Google Scholar or Semantic Scholar
C: conference / J: journal / P: preprint
*: equal contribution

[J3] Estimating Simplet Counts via Sampling

  • Hyunju Kim*, Heechan Moon*, Fanchen Bu, Jihoon Ko, and Kijung Shin
  • The VLDB Journal [ paper Coming soon | code | bib Coming soon ]
  • (Journal Extension of [C3])

[C13] DiffIM: Differentiable Influence Minimization with Surrogate Modeling and Continuous Relaxation

  • Junghun Lee, Hyunju Kim, Fanchen Bu, Jihoon Ko, Kijung Shin
  • AAAI 2025 [ paper | code Coming soon | bib ]

[C12] On Measuring Unnoticeability of Graph Adversarial Attacks: Observations, New Measure, and Applications

  • Hyeonsoo Jo*, Hyunjin Hwang*, Fanchen Bu, Soo Yong Lee, Chanyoung Park, and Kijung Shin

[C11] Rethinking Reconstruction-based Graph-Level Anomaly Detection: Limitations and a Simple Remedy

  • Sunwoo Kim, Soo Yong Lee, Fanchen Bu, Shinhwan Kang, Kyungho Kim, Jaemin Yoo, and Kijung Shin

[P3] RL4CO: an Extensive Reinforcement Learning for Combinatorial Optimization Benchmark

  • Federico Berto*, Chuanbo Hua*, Junyoung Park*, Laurin Luttmann*, Yining Ma, Fanchen Bu, Jiarui Wang, Haoran Ye,

[P2] Exploring Edge Probability Graph Models Beyond Edge Independency: Concepts, Analyses, and Algorithms

  • Fanchen Bu, Ruochen Yang, Paul Bogdan, and Kijung Shin

[C10] Tackling Prevalent Conditions in Unsupervised Combinatorial Optimization: Cardinality, Minimum, Covering, and More

  • Fanchen Bu, Hyeonsoo Jo, Soo Yong Lee, Sungsoo Ahn, and Kijung Shin

[C9] Feature Distribution on Graph Topology Mediates the Effect of Graph Convolution: Homophily Perspective

  • Soo Yong Lee, Sunwoo Kim, Fanchen Bu, Jaemin Yoo, Jiliang Tang, and Kijung Shin

[P1] A Survey on Hypergraph Mining: Patterns, Tools, and Generators

  • Geon Lee*, Fanchen Bu*, Tina Eliassi-Rad, and Kijung Shin

[C8] HypeBoy: Generative Self-Supervised Representation Learning on Hypergraphs

  • Sunwoo Kim, Shinhwan Kang, Fanchen Bu, Soo Yong Lee, Jaemin Yoo, and Kijung Shin

[C7] Robust Graph Clustering via Meta Weighting for Noisy Graphs

  • Hyeonsoo Jo, Fanchen Bu, and Kijung Shin

[C6] On Improving the Cohesiveness of Graphs by Merging Nodes: Formulation, Analysis, and Algorithms

  • Fanchen Bu and Kijung Shin

[C5] How Transitive Are Real-World Group Interactions? - Measurement and Reproduction

  • Sunwoo Kim, Fanchen Bu, Minyoung Choe, Jaemin Yoo, and Kijung Shin

[J2] Interplay between Topology and Edge Weights in Real-World Graphs: Concepts, Patterns, and an Algorithm

  • Fanchen Bu, Shinhwan Kang, and Kijung Shin
  • (ECML PKDD 2023 Journal Track)

[C4] Towards Deep Attention in Graph Neural Networks: Problems and Remedies

  • Soo Yong Lee, Fanchen Bu, Jaemin Yoo, and Kijung Shin

[C3] Characterization of Simplicial Complexes by Counting Simplets Beyond Four Nodes

  • Hyunju Kim, Jihoon Ko, Fanchen Bu, and Kijung Shin

[J1] Hypercore Decomposition for Non-Fragile Hyperedges: Concepts, Algorithms, Observations, and Applications

  • Fanchen Bu, Geon Lee, and Kijung Shin
  • (ECML PKDD 2023 Journal Track)

[C2] Feedback Gradient Descent: Efficient and Stable Optimization with Orthogonality for DNNs

  • Fanchen Bu and Dong Eui Chang

[C1] Double Prioritized State Recycled Experience Replay

  • Fanchen Bu and Dong Eui Chang

Languages

Chinese

Native

English

Advanced

  • IELTS 7.5
  • GRE 333

Korean

Intermediate

Academic Services

[J4] International Journal of Machine Learning and Cybernetics

  • Reviewer: 2024

[J3] IEEE Transactions on Network Science and Engineering

  • Reviewer: 2024

[J2] The Journal of Supercomputing

  • Reviewer: 2024

[J1] Big Data Research

  • Reviewer: 2024

[C6] The Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD)

  • Workshop Organizer: 2025 (GLFM)

[C5] The ACM Web Conference (TheWebConf/WWW)

  • Reviewer: 2025

[C4] Learning on Graphs Conference (LoG)

  • Reviewer: 2024
  • Note: Selected as one of the top reviewers (32 in total) in LoG 2024

[C3] European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD)

  • Reviewer: 2024

[C2] Asian Conference on Machine Learning (ACML)

  • Reviewer: 2024

[C1] ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD)

  • Reviewer: 2024 - 2025
  • Note: Selected as one of the outstanding reviewers (top 10%) in KDD 2025 (August Cycle)

Favorite Music

The links are on Youtube Music
  1. Chopin's Nocturnes by Arthur Rubinstein / Claudio Arrau / Đặng Thái Sơn
  1. Chopin's Piano Concertos by Arthur Rubinstein / Yundi Li [李雲迪] / Seong-Jin Cho [조성진]
  1. Chinese traditional music by National Cinema Symphony Orchestra [中國廣播電影交響樂團]
  1. Mandopop by David Tao [陶喆] / Jay Chou [周傑倫] / Ronghao Li [李榮浩]
  1. Cantopop by Hins Cheung [張敬軒] / Janice Vidal [衛蘭]
  1. Korean ballades by Si-Kyung Sung [성시경] / Kwang-Seok Kim [김광석]
  1. Japanese Enka by Teresa Teng [テレサ・テン; 鄧麗君] / Shinichi Mori [森 進一] / Masako Mori [森 昌子]
  1. Folk music by John Denver

Useful Tools and Websites

  1. DeskPins: a tool for making a window topmost
  1. Microsoft PowerToys: a set of utilities for power users to tune and streamline their Windows experience for greater productivity
  1. TreeSize Free: disk space analyzer
  1. Everything: a tool for searching files
  1. Acronymify: automatic acronym generator
  1. Linggle: language reference search engines
  1. Academic Biblical Studies (Reddit): a subreddit for discussion of early Judaism and Christianity in a scholarly context
  1. Chord Chart Studio: an online chord chart rendering platform
  1. Two Youtube channels for classical music analysis and appreciation:
    1. gregniemczuk_official: Polish pianist Grzegorz Niemczuk
    1. Sonata Secrets: Swedish pianist Henrik Kilhamn
  1. MangoldProject: a Youtube channel teaching piano/keyboard skills with music theory

Miscellanea

  1. I am active on Mathematics Stack Exchange (a math Q&A cite)
  1. My MBTI is Assertive Executive (ESTJ-A)
  1. I play League of Legends (since Season 2, year 2011)
    1. My highest tier was solo-rank Master in both Chinese and Korean servers
  1. I play Japanese riichi mahjong
    1. My highest tier was 4-Dan (四段) in Tenhou (天鳳) and Master (雀豪) in Mahjong Soul (雀魂)