Avatar

Mingyuan "William" Zhang
张明远

About Me

I am a Ph.D. student researcher focusing on Machine Learning and Artificial Intelligence, having the honor to be advised by Prof. Shivani Agarwal at the University of Pennsylvania. Before joining Penn, I received B.S. degree from the University of Michigan in 2018 with four majors: (Honors) Mathematics, (Honors) Statistics, Computer Science and Data Science.

"Think big, start small, learn fast.
Seek progress, not perfection."

Résumé

Publications

Google Scholar

Semantic Scholar

GitHub

[email protected]


Education

University of Pennsylvania
UPenn logo
Ph.D. in Computer and Information Science
2018 - Present
University of Michigan
UM logo
B.S. in Honors Mathematics, Honors Statistics, Computer Science, and Data Science
2013 - 2018

Academic Service

Reviewer
• NeurIPS (2021, 2022, 2023, 2024)
• ICLR (2022, 2023)
• AISTATS (2024)
• Journal of Machine Learning Research
• Transactions on Pattern Analysis and Machine Intelligence

Teaching

University of Pennsylvania
• Head Teaching Assistant for CIS 520, a graduate level machine learning course.
Spring 2020, Spring 2021, Spring 2022
University of Michigan
• Grader for various linear algebra and probability courses.
2015 - 2018
• Tutor for MATH 217, an introductory linear algebra course.
2015

Courses

Graduate level:
Real Analysis (A), Probability Theory (A), Discrete Stochastic Processes (A), Numerical Linear Algebra (A+), Combinatorial Theory (A+), Complex Variables (A), Applied Functional Analysis (A), Nonlinear Programming (A+), Statistical Inference (A), Linear Models (A), Analysis of Multivariate and Categorical Data (A), Statistical Learning (A), Time Series Analysis (A-), Machine Learning (A), Information Theory (A+), Statistical Signal Processing (A).
Undergraduate level:
Intermediate Microeconomics Theory (A+), Intermediate Macroeconomics Theory (A), Game Theory (A+), Theoretical Statistics (A+), Statistical Computing Methods (A+), Numerical Methods (A+), Programming and Data Structures (A+), Data Structures and Algorithms (A+), Algorithms (A), Randomized Algorithms (A+), Database Management Systems (A), Computer Vision (A), Information Retrieval (A).

Research

Multiclass Classification
Multiclass classification (which includes binary classification) is a classic supervised machine learning task. We study how to design good multiclass learning algorithms for general losses in various settings including the standard setting, learning with a restricted function class, and learning from noisy labels.

[5] Multiclass Learning from Noisy Labels for Non-decomposable Performance Measures.
Mingyuan Zhang, Shivani Agarwal.
In Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS), 2024.
Paper Link

[3] Learning from Noisy Labels with No Change to the Training Process.
Mingyuan Zhang, Jane Lee, Shivani Agarwal.
In Proceedings of the 38th International Conference on Machine Learning (ICML), 2021.
Paper Link

[2] Bayes Consistency vs. H-Consistency: The Interplay between Surrogate Loss Functions and the Scoring Function Class.
Mingyuan Zhang, Shivani Agarwal.
In Advances in Neural Information Processing Systems (NeurIPS), 2020.
Spotlight paper.
Paper Link

Multiclass and multi-label learning with general losses: What is the right output coding and decoding?
Harish G. Ramaswamy, Mingyuan Zhang, Balaji S. Babu, Shivani Agarwal, Ambuj Tewari, Robert C. Williamson.
In preparation.


Learning from Noisy Labels and Weakly Supervised Learning
Noisy labels can occur due to various reasons such as errors in data collection, human error in annotating data, or mislabeling due to subjective or ambiguous definitions. The main challenge in learning from noisy labels is to design algorithms that can learn good classifiers, despite being given noisy training data. We study how to design good algorithms to learn from noisy labels for multiclass and multi-label classification problems.
We are also interested in a more general learning scheme beyond learning from noisy labels, specifically, weakly supervised learning. Our focus is on learning from missing or partial labels, and transfer learning.

Consistent Multi‑Label Learning from Noisy Labels.
Mingyuan Zhang, Shivani Agarwal.
Under review, 2024.

[5] Multiclass Learning from Noisy Labels for Non-decomposable Performance Measures.
Mingyuan Zhang, Shivani Agarwal.
In Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS), 2024.
Paper Link

[4] Foreseeing the Benefits of Incidental Supervision.
Hangfeng He, Mingyuan Zhang, Qiang Ning, Dan Roth.
In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2021.
Oral paper.
Paper Link

[3] Learning from Noisy Labels with No Change to the Training Process.
Mingyuan Zhang, Jane Lee, Shivani Agarwal.
In Proceedings of the 38th International Conference on Machine Learning (ICML), 2021.
Paper Link


Multi-Label Classification and Label Ranking
In multi-label classification, each instance can be associated with multiple labels (or tags) simultaneously. A good example is image tagging, where several tags can be active in the same image. We study how to design good multi-label learning algorithms for general multi-label losses (including Hamming Loss, Precision, Recall and F-measure) in various settings including the standard setting, learning from noisy labels, and learning from partial/missing labels.
Label ranking is a prediction task where the goal is to map instances to rankings over a finite set of predefined labels (or tags). We study the design of effective label ranking learning algorithms for a range of label ranking losses. These include Pairwise Loss, Discounted Cumulative Gain, and Precision, and our study applies to various settings such as the standard and online settings.

Consistent Multi‑Label Learning from Noisy Labels.
Mingyuan Zhang, Shivani Agarwal.
Under review, 2024.

On the Minimax Regret in Online Ranking with Top-k Feedback.
Mingyuan Zhang, Ambuj Tewari.
Preprint, under review, 2023.
Paper Link

[1] Convex Calibrated Surrogates for the Multi-Label F-Measure.
Mingyuan Zhang, Harish G. Ramaswamy, Shivani Agarwal.
In Proceedings of the 37th International Conference on Machine Learning (ICML), 2020.
Paper Link

Multiclass and multi-label learning with general losses: What is the right output coding and decoding?
Harish G. Ramaswamy, Mingyuan Zhang, Balaji S. Babu, Shivani Agarwal, Ambuj Tewari, Robert C. Williamson.
In preparation.


Non‑decomposable Performance Measures
Unlike 0-1 loss (accuracy) or cost-sensitive losses, non-decomposable performance measures cannot be expressed as the expectation or sum of a loss on individual examples; such performance measures are defined by general (usually nonlinear) functions of the confusion matrix of a classifier. Examples include Micro F1 score, Jaccard measure, H-mean, G-mean, Q-mean, AUC-ROC, and AUC-PR. We study how to design learning algorithms to optimize for non-decomposable performance measures.

[5] Multiclass Learning from Noisy Labels for Non-decomposable Performance Measures.
Mingyuan Zhang, Shivani Agarwal.
In Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS), 2024.
Paper Link