Hi - I’m Neil, although I generally use my formal name Ziyu Xu (许子俞) on my research.
I’m a PhD student in Statistics and Data Science at CMU.
I’m fortunate enough to be advised by Aaditya Ramdas. I also collaborate with Ruodu Wang.
I’ve recently been working on multiple hypothesis testing and post-selection inference, and I’m interested in the intersection of statistical and game theory.
I am grateful to Justin Khim, Pradeep Ravikumar, and David P. Woodruff for showing me the ropes of how to do theory research.
In my undergrad, I had a stint of working on natural language processing applications, and had the fortune of working with Zhou Yu on dialog agents and Graham Neubig on machine translation.
Email: ecks zee why at cmu dot edu
Preprints
-
Online generalizations of the e-BH and BH procedure
Lasse Fischer, Ziyu Xu, Aaditya Ramdas
2024 [arXiv 2407.20683]
-
More powerful multiple testing under dependence via randomization
Ziyu Xu, Aaditya Ramdas
2023 [arXiv 2305.11126] [code]
Publications
-
Active, anytime-valid risk controlling prediction sets
Ziyu Xu, Nikos Karampatziakis, Paul Mineiro
NeurIPS 2024 [arXiv 2406.10490] [code]
-
Post-selection inference for e-value based confidence intervals
Ziyu Xu, Ruodu Wang, Aaditya Ramdas
Electronic Journal of Statistics 2024 [arXiv 2203.12572] [poster] [talk] [slides]
Runner-up Poster Prize @ MCP 2022
-
Online multiple testing with e-values
Ziyu Xu, Aaditya Ramdas
AISTATS 2024 [arXiv 2311.06412] [proceedings] [poster] [slides] [code]
-
Risk-limiting Financial Audits via Weighted Sampling without Replacement
Shubhanshu Shekhar, Ziyu Xu, Zachary C. Lipton, Pierre J. Liang, Aaditya Ramdas
UAI 2023 [arXiv 2305.06884] [proceedings] [code]
-
Memory Bounds for the Experts Problem
Vaidehi Srinivas, David P. Woodruff, Ziyu Xu, Samson Zhou
STOC 2022 [arXiv 2204.09837] [proceedings] [Samson's talk]
-
A unified framework for bandit multiple testing
Ziyu Xu, Ruodu Wang, Aaditya Ramdas
NeurIPS 2021 [arXiv 2107.07322] [proceedings] [poster] [talk] [slides] [code]
-
Dynamic Algorithms for Online Multiple Testing
Ziyu Xu, Aaditya Ramdas
Math. and Sci. ML 2021 [arXiv 2010.13953] [proceedings] [talk] [slides] [code]
-
Class-Weighted Classification: Trade-offs and Robust Approaches
Ziyu Xu, Chen Dan, Justin Khim, Pradeep Ravikumar
ICML 2020 [arXiv 2005.12914] [proceedings] [talk] [code]
-
Strategy and Policy Learning for Non-Task-Oriented Conversational Systems
Zhou Yu, Ziyu Xu, Alan W Black, Alexander Rudnicky
SIGDIAL 2016 [proceedings]
-
Chatbot Evaluation and Database Expansion via Crowdsourcing
Zhou Yu, Ziyu Xu, Alan W Black, Alexander Rudnicky
RE-WOCHAT workshop of LREC 2016 [proceedings]