Publications

Journal Publications

1. Information-theoretic Characterizations of Generalization Error for the Gibbs Algorithm
   G. Aminian*, Y. Bu*, L. Toni, M. R. Rodrigues, G. W. Wornell (* equal contribution)
   IEEE Transactions on Information Theory, vol. 70, no. 1, pp. 632-655, Jan. 2024.
   

2. A Maximal Correlation Framework for Fair Machine Learning
   Y. Bu*, J. K. Lee*, P. Sattigeri, R. Panda, G. W. Wornell, L. Karlinsky, R. S. Feris (* equal contribution)
   Entropy 24, no. 4, pp. 461, Mar. 2022.
   

3. Population Risk Improvement with Model Compression: An Information-Theoretic Approach
   Y. Bu, W. Gao, S. Zou, V. V. Veeravalli.
   Entropy 23, no. 10, pp. 1255, Sept. 2021.
   

4. Tightening Mutual Information Based Bounds on Generalization Error
   Y. Bu, S. Zou, V. V. Veeravalli.
   IEEE Journal on Selected Areas in Information Theory, vol. 1, pp. 121 - 130, May 2020.
   

5. Adaptive Sequential Machine Learning
   C. Wilson, Y. Bu, V. V. Veeravalli.
   Sequential Analysis, vol. 38, no. 4, Oct. 2019. (16th Abraham Wald Prize)
   

6. Linear-Complexity Exponentially-Consistent Tests for Universal Outlying Sequence Detection
   Y. Bu, S. Zou, V. V. Veeravalli.
   IEEE Transactions on Signal Processing, vol. 67, no. 8, pp. 2115-2128, Apr. 2019.
   

7. Estimation of KL Divergence: Optimal Minimax Rate
   Y. Bu*, S. Zou*, Y. Liang, V. V. Veeravalli. (*equal contribution),
   IEEE Transactions on Information Theory, vol. 64, no. 4, pp. 2648-2674, Apr. 2018.
   Code available at Github

Conference and Workshop Publications

2024

• Class-wise Generalization Error: an Information-Theoretic Analysis
  F. Laakom, Y. Bu, M. Gabbouj
  Short version at ICML workshop on PAC-Bayes Meets Interactive Learning, Honolulu, HI, Jul. 2023.

• Are Uncertainty Quantification Capabilities of Evidential Deep Learning a Mirage?
  J. J. Ryu, M. Shen, S. Ghosh, Y. Bu, P. Sattigeri, S. Das, G. W. Wornell
  to apear in, Conference on Neural Information Processing Systems (NeurIPS), Dec. 2024. (acceptance rate: 26%)
  

• Information-theoretic Analysis of the Gibbs Algorithm: An Individual Sample Approach
  Y. Zhu, Y. Bu
  to appear in, IEEE Information Theory Workshop (ITW), Shenzhen, China, Nov. 2024.

• SAUC: Sparsity-Aware Uncertainty Calibration for Spatiotemporal Prediction with Graph Neural Networks
  D. Zhuang, Y. Bu, G. Wang, S. Wang, J. Zhao
  to appear in, International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL), Atlanta, GA, Oct. 2024
  Short version at NeurIPS Workshop on Temporal Graph Learning (TGL), New Orleans, LA, Dec. 2023. (Spotlight talk)

• Learning Orthonormal Features in Self-Supervised Learning using Functional Maximal Correlation
  B. Hu, Y. Bu, J. C. Príncipe
  to appear in, IEEE International Conference on Image Processing (ICIP), Abu Dhabi, United Arab Emirates, Oct. 2024.
  Short version at NeurIPS Workshop on Self-Supervised Learning - Theory and Practice, New Orleans, LA, Dec. 2023. (Spotlight talk)

• An Algorithm for Computing the Capacity of Symmetrized KL Information for Discrete Channels
  H. Chen, G. Aminian, Y. Bu
  to appear in, Allerton Conference on Communication, Control, and Computing (Allerton), Urbana, IL, Sep. 2024

• Information-Theoretic Opacity-Enforcement in Markov Decision Processes
  C. Shi, Y. Bu, J. Fu
  International Joint Conference on Artificial Intelligence (IJCAI), Jeju, Korea, Aug. 2024. (acceptance rate: 15%)
  

• Adaptive Text Watermark for Large Language Models
  Y. Liu, Y. Bu
  International Conference on Machine Learning (ICML), Vienna, Austria, Jul. 2024. (acceptance rate: 28%)
  Code available at Github

• Operator SVD with Neural Networks via Nested Low-Rank Approximation
  J. J. Ryu, X. Xu, H. SM Erol, Y. Bu, L. Zheng, G. W. Wornell
  International Conference on Machine Learning (ICML), Vienna, Austria, Jul. 2024. (acceptance rate: 28%)
  Short version at NeurIPS Workshop on Machine Learning and the Physical Sciences, New Orleans, LA, Dec. 2023.
  Code available at Github

• Towards Optimal Inverse Temperature in the Gibbs Algorithm
  Y. Bu
  IEEE International Symposium on Information Theory (ISIT), Athens, Greece, Jul. 2024.
  

• Group Fairness with Uncertainty in Sensitive Attributes
  A. Shah, M. Shen, J. J. Ryu, S. Das, P. Sattigeri, Y. Bu, G. W. Wornell
  IEEE International Symposium on Information Theory (ISIT), Athens, Greece, Jul. 2024.
  Full version at ICML workshop on Spurious Correlations, Invariance and Stability, Honolulu, HI, Jul. 2023.

• Gibbs-Based Information Criteria and the Over-Parameterized Regime
  H. Chen, G. W. Wornell, Y. Bu
  International Conference on Artificial Intelligence and Statistics (AISTATS), Valencia, Spain, May 2024. (acceptance rate: 28%)
  Short version at NeurIPS Workshop on Mathematics of Modern Machine Learning (M3L), New Orleans, LA, Dec. 2023.

2023

• On Balancing Bias and Variance in Unsupervised Multi-Source-Free Domain Adaptation
  M. Shen, Y. Bu, G. W. Wornell
  International Conference on Machine Learning (ICML), Honolulu, HI, Jul. 2023. (acceptance rate: 28%)

• On the Generalization Error of Meta Learning for the Gibbs Algorithm
  Y. Bu, H. V. Tetali, G. Aminian, M. R. Rodrigues, G. W. Wornell
  IEEE International Symposium on Information Theory (ISIT), Taipei, Jun. 2023.

• A Bilateral Bound on the Mean Squared Error for Estimation in Model Mismatch
  A. Weiss, A. Lancho, Y. Bu, G. W. Wornell
  IEEE International Symposium on Information Theory (ISIT), Taipei, Jun. 2023.

• Reliable Gradient-free and Likelihood-free Prompt Tuning
  M. Shen, S. S. Ghosh, P. Sattigeri, S. Das, Y. Bu, G. W. Wornell
  Findings of the European Chapter of the Association for Computational Linguistics (EACL), Dubrovnik, Croatia, May 2023.

• How Does Pseudo-Labeling Affect the Generalization Error of the Semi-Supervised Gibbs Algorithm?
  H. He, G. Aminian, Y. Bu, M. R. Rodrigues, V. YF Tan
  International Conference on Artificial Intelligence and Statistics (AISTATS), Valencia, Spain, Apr. 2023. (acceptance rate: 29%)

• Post-hoc Uncertainty Learning using a Dirichlet Meta-Model
  M. Shen, Y. Bu, P. Sattigeri, S. S. Ghosh, S. Das, G. W. Wornell
  AAAI Conference on Artificial Intelligence (AAAI), Washington DC, Feb. 2023. (acceptance rate: 19%)
  Code available at Github

2022

• Data-Driven Blind Synchronization and Interference Rejection for Digital Communication Signals
  A. Lancho, A. Weiss, G. C. Lee, J. Tang, Y. Bu, Y. Polyanskiy, G. W. Wornell
  IEEE Global Communications Conference, Rio de Janeiro, Brazil, Dec. 2022.

• Exploiting Temporal Structures of Cyclostationary Signals for Data-Driven Single-Channel Source Separation
  G. C. Lee, A. Weiss, A. Lancho, J. Tang, Y. Bu, Y. Polyanskiy, G. W. Wornell
  IEEE International Workshop on Machine Learning for Signal Processing, Xi'an, China, Aug. 2022. (Best Student Paper)

• Selective Regression under Fairness Criteria
  A. Shah*, Y. Bu*, J. K. Lee, P. Sattigeri, R. Panda, S. Das, G. W. Wornell (* equal contribution)
  International Conference on Machine Learning (ICML), Baltimore, MD, Jul. 2022. (acceptance rate: 22%)

• Tighter Expected Generalization Error Bounds via Convexity of Information Measures
  G. Aminian*, Y. Bu*, G. W. Wornell, M. R. Rodrigues (* equal contribution)
  IEEE International Symposium on Information Theory (ISIT), Espoo, Finland, Jun. 2022.

• A Maximal Correlation Approach to Imposing Fairness in Machine Learning
  Y. Bu*, J. K. Lee*, P. Sattigeri, R. Panda, G. W. Wornell, L. Karlinsky, R. S. Feris (*equal contribution)
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Singapore, May 2022.

• Characterizing and Understanding the Generalization Error of Transfer Learning with Gibbs Algorithm
  Y. Bu*, G. Aminian*, L. Toni, M. R. Rodrigues, G. W. Wornell (* equal contribution)
  International Conference on Artificial Intelligence and Statistics (AISTATS), virtual, Mar. 2022. (acceptance rate: 29%)

Before 2022

• An Exact Characterization of the Generalization Error for the Gibbs Algorithm
  G. Aminian*, Y. Bu*, L. Toni, M. R. Rodrigues, G. W. Wornell (* equal contribution)
  Conference on Neural Information Processing Systems (NeurIPS), virtual, Dec. 2021. (acceptance rate: 26%)
  Short version at ICML Workshop on Information-Theoretic Methods for Rigorous, Responsible, and Reliable Machine Learning, Jul. 2021.

• Fair Selective Classification via Sufficiency
  Y. Bu*, J. K. Lee*, D. Rajan, P. Sattigeri, R. Panda, S. Das, G. W. Wornell. (* equal contribution)
  International Conference on Machine Learning (ICML), virtual, Jul. 2021. (Oral, Top 3%)

• SDP Methods for Sensitivity-Constrained Privacy Funnel and Information Bottleneck Problems
  Y. Bu, T. Wang, G. W. Wornell
  IEEE International Symposium on Information Theory (ISIT), virtual, Jul. 2021.

• Information-theoretic Understanding of Population Risk Improvement with Model Compression
  Y. Bu, W. Gao, S. Zou, V. V. Veeravalli
  AAAI Conference on Artificial Intelligence (AAAI), New York, NY, Feb. 2020. (acceptance rate: 21%)

• Active Learning in Recommendation Systems with Multi-level User Preferences
  Y. Bu, K. Small
  AAAI Workshop on Interactive and Conversational Recommendation Systems (WICRS), New York, NY, Feb. 2020.

• Active and Adaptive Sequential Learning with Per Time-step Excess Risk Guarantees
  Y. Bu, J. Lu, V. V. Veeravalli
  IEEE Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, Nov. 2019.
  Full version available on arXiv

• Tightening Mutual Information Based Bounds on Generalization Error
  Y. Bu, S. Zou, V. V. Veeravalli
  IEEE International Symposium on Information Theory (ISIT), Paris, France, Jul. 2019.

• Model Change Detection with Application to Machine Learning
  Y. Bu, J. Lu, V. V. Veeravalli
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, UK, May 2019.

• Linear-Complexity Exponentially-Consistent Tests for Universal Outlying Sequence Detection
  Y. Bu, S. Zou, V. V. Veeravalli
  IEEE International Symposium on Information Theory (ISIT), Aachen, Germany, Jun. 2017.

• Estimation of KL Divergence between Large-Alphabet Distributions
  Y. Bu, S. Zou, Y. Liang, V. V. Veeravalli
  IEEE International Symposium on Information Theory (ISIT), Barcelona, Spain, Jul. 2016.

• Universal Outlying Sequence Detection for Continuous Observations
  Y. Bu, S. Zou, Y. Liang, V. V. Veeravalli
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Shanghai, China, Mar. 2016.