IEEE Trans. Information Theory 2023

212 papers
In 2023, research in IEEE Transactions on Information Theory focused significantly on network analysis and community detection. Key studies addressed degree heterogeneity, the management of a growing number of clusters, and degree-agnostic spectral clustering for heterogeneous networks. There was a notable emphasis on hypergraph learning, specifically regarding the preservation of higher-order structures and the consistency of spectral clustering in sparse models. Other contributions explored universal graph compression and the development of logarithmically larger deletion codes. The year's edition also featured extensive work on statistical inference and machine learning theory. Prominent topics included anytime-valid inference, confidence sequences derived from online learning regret, and tight concentration inequalities for unbounded random variables. Researchers such as Junjie Ma, Xiuyuan Cheng, and Ben Graham investigated optimal transport limits in adversarial learning and phase transitions in robust classification. Additionally, papers examined distributed mean estimation with side information and Gaussian mixture reduction using composite transportation divergence. These works collectively advanced the understanding of fundamental limits in crowdsourcing and sequential testing.

Top Research Directions

Research directions most represented by IEEE Trans. Information Theory 2023 papers.

Asymptotically Optimal Quickest Change Detection 3
The All or Nothing Phenomenon in High Dimensional Inference 3
Overcoming the Gaussian Assumption in High-Dimensional Learning 3
The Prevalence Threshold in Group Testing Efficiency 2
Exploiting Statistical Correlations in Group Testing 2
Quantifying Generalization via Information Leakage 2
Distributional Smoothness Constraints in Statistical Learning Theory 2
Mitigating Biochemical Noise in DNA Data Storage 2
Channel Symmetrizability in Adversarial Communication 2
Anytime Valid Inference via Testing by Betting 2
Minimizing Finite Field Complexity in Error Correcting Codes 2
The Sample Complexity Gap in Sparse Recovery 2
The Statistical Computational Gap in High Dimensions 2
Noise Robust Sparse Recovery 2
Mitigating the Curse of Dimensionality in Density Estimation 2

Top Authors

AuthorPapers
Vincent Y. F. Tan6
Antonia Wachter-Zeh5
Alexandre Graell i Amat4
Ayfer Ozgur4
Alex Dytso3
Cheuk Ting Li3
Yao Xie3
Yihan Zhang3
Ziv Goldfeld3
Alfred O. Hero2
Aryeh Kontorovich2
Ashish J Khisti2
Bingchen Qian2
Florent Krzakala2
Hamed Hassani2
Jonathan Scarlett2
Liyan Xie2
Michael Ng2
P. N. Karthik2
Qiaosheng Zhang2

Impactful Papers

Papers from IEEE Trans. Information Theory 2023 that shaped research clusters.

Influential

Papers

Title
Uncertainty Quantification for Nonconvex Tensor Completion: Confidence Intervals, Heteroscedasticity and Optimality
Undecidability of Network Coding, Conditional Information Inequalities, and Conditional Independence Implication
Uniform Exact Reconstruction of Sparse Signals and Low-Rank Matrices From Phase-Only Measurements
Unitarity Estimation for Quantum Channels
Universal Graph Compression: Stochastic Block Models
Universality Laws for High-Dimensional Learning With Random Features
Unsourced Multiple Access With Random User Activity
Virtual Quantum Markov Chains
WildWood: A New Random Forest Algorithm
Window-Limited CUSUM for Sequential Change Detection
Wyner-Ziv Estimators for Distributed Mean Estimation With Side Information and Optimization
Zero-Error Communication Over Adversarial MACs

« prev Page 5/5