Yingbin Liang
Engineering · The Ohio State University
Publications
395
Citations
8,054
Est. group size
~2
Recurring co-author estimate
Active years
29
Publishing since 1998
Yingbin Liang's work focuses on the theoretical foundations of machine learning, especially reinforcement learning (a method where an algorithm learns by trial and error to make good decisions), and the mathematical analysis of how models like transformers and diffusion models learn and converge during training. This research is aimed at understanding when and why learning algorithms work, including under uncertainty, partial information, and imperfect data. The publication record also includes many applied deep learning projects (e.g., image classification, fraud detection, medical text analysis), suggesting collaboration across a range of applied topics alongside the core theoretical agenda.
Publication output has fluctuated over the last decade, dipping around 2021-2022 and rebounding in 2024, with a notable drop in 2025-2026 that may reflect incomplete recent data.
Generated by claude-sonnet-5 from public bibliographic data · Jul 20, 2026
- Convergence Guarantees for Time-Inhomogeneous Uniform-Rate Discrete Diffusion Models
Entropy · 2026
- Reinforcement Learning With Partial Online State Information in POMDPs: Regret Bounds and Limits
IEEE Transactions on Information Theory · 2026
- Reinforcement Learning from Multi-Source Imperfect Preferences: Best-of-Both-Regimes Regret
arXiv (Cornell University) · 2026
- How Transformers Learn Regular Language Recognition: A Theoretical Study on Training Dynamics and Implicit Bias
arXiv (Cornell University) · 2025
- Multi-head Transformers Provably Learn Symbolic Multi-step Reasoning via Gradient Descent
arXiv (Cornell University) · 2025
- Contrastive and Variational Approaches in Self-Supervised Learning for Complex Data Mining
arXiv (Cornell University) · 2025
- Robust Offline Reinforcement Learning for Non-Markovian Decision Processes
IEEE Transactions on Information Theory · 2025
- Influence of process parameters on microscopic morphology and mechanical properties of 3D printed continuous carbon fiber reinforced polyamide6 composites
Journal of Materials Research and Technology · 2025
- In-Context Learning with Representations: Contextual Generalization of Trained Transformers
2024
- Editorial Data, Physics, and Life Through the Lens of Information Theory
IEEE Journal on Selected Areas in Information Theory · 2024
- Integrative Analysis of Financial Market Sentiment Using CNN and GRU for\n Risk Prediction and Alert Systems
arXiv (Cornell University) · 2024
- Robust Offline Reinforcement Learning for Non-Markovian Decision Processes
arXiv (Cornell University) · 2024
- Non-asymptotic Convergence of Training Transformers for Next-token Prediction
arXiv (Cornell University) · 2024
- Dynamic Fraud Detection: Integrating Reinforcement Learning into Graph Neural Networks
arXiv (Cornell University) · 2024
- Suppression of Edge Localized Modes in ITER Baseline Scenario in EAST using Edge Localized Magnetic Perturbations
arXiv (Cornell University) · 2024
- arXiv (Cornell University)×137
- IEEE Transactions on Information Theory×15
- Zenodo (CERN European Organization for Nuclear Research)×15
- Neural Information Processing Systems×5
- IEEE Transactions on Signal Processing×5
- Mohsen Heidari
Engineering · Indiana University
- Aylin Yener
Engineering · The Ohio State University
- Ahmed Bendary
Computer Science · The Ohio State University
- Chih-Chun Wang
Computer Science · Purdue University West Lafayette
- Jonathan Ponniah
Computer Science · Indiana University
This profile was generated automatically from public scholarly data (OpenAlex). Group size and activity levels are estimates derived from co-authorship patterns.
Last updated Jul 19, 2026.
Claim or correct this profile