CV

Education

• B.S., in Electrical Engineering, Hangzhou Dianzi University, 2016
• M.S., in Computer Science, University of Wisconsin - Madison, 2018 (expected)
• Ph.D, in Computer Science, University of Wisconsin - Madison, 2022 (expected)

Work experience

• Fall 2017: Research Assistant
  • ECE Department of UW - Madison
  • Duties included: Research on distributed machine learning
  • Supervisor: Prof. Dimitris Papailiopoulos
• Spring & Summer 2017: Research Assistant
  • CS Department of UW - Madison
  • Duties included: Research on robotics
  • Supervisor: Prof. Micheal Gleicher

Skills

• Programming Languages: C/C++, Python, MATLAB, Java, Scala
• Operating Systems: Linux, Windows, MacOSX
• Deep Learning Frameworks: TensorFlow, MXNet, PyTorch
• Distributed Systems: Spark
• Robotics: ROS, Moveit
• Hardware: RF Circuit Design, PCB Design, Embedded Systems

Honors & Awards

• Student Travel Award, NIPS 2018
  • Dec 2018, issued by Thirty-second Conference on Neural Information Processing Systems (NIPS 2018)
• Student Travel Award, ICML 2018
  • May 2018, issued by Thirty-fifth International Conference on Machine Learning (ICML 2018)
• Huawei Scholarship
  • Nov 2015, issued by Huawei Technologies Co., Ltd.
• National Scholarship of China
  • Oct 2015, issued by Ministry of Education of the People’s Republic of China

Publications

Recognizing Actions during Tactile Manipulations through Force Sensing

Draco: Robust Distributed Training against Adversaries

DRACO: Robust Distributed Training via Redundant Gradients

ATOMO: Communication-efficient Learning via Atomic Sparsification

The Effect of Network Width on the Performance of Large-batch Training

ErasureHead: Distributed Gradient Descent without Delays Using Approximate Gradient Coding

Demonstration of Nimbus: Model-based Pricing for Machine Learning in a Data Marketplace

DETOX: A Redundancy-based Framework for Faster and More Robust Gradient Aggregation

Federated Learning with Matched Averaging

Attack of the Tails: Yes, You Really Can Backdoor Federated Learning

FedML: A Research Library and Benchmark for Federated Machine Learning

Accordion: Adaptive Gradient Communication via Critical Learning Regime Identification

Pufferfish: Communication-efficient Models At No Extra Cost

On the Utility of Gradient Compression in Distributed Training Systems

AMP: Automatically Finding Model Parallel Strategies with Heterogeneity Awareness

Rare Gems: Finding Lottery Tickets at Initialization

Efficient Federated Learning on Knowledge Graphs via Privacy-preserving Relation Embedding Aggregation

Federated Learning as Variational Inference: A Scalable Expectation Propagation Approach

MPCFormer: fast, performant and private Transformer inference with MPC

Cuttlefish: Low-rank Model Training without All The Tuning

FedNAR: Federated Optimization with Normalized Annealing Regularization

Fusing Models with Complementary Expertise

Does compressing activations help model parallel training?

RedCoast: A Lightweight Tool to Automate Distributed Training of LLMs on Any GPU/TPUs

Crystal: Illuminating LLM Abilities on Language and Code

LLM360: Towards Fully Transparent Open-Source LLMs

Maestro: Uncovering Low-Rank Structures via Trainable Decomposition

TrustLLM: Trustworthiness in Large Language Models

FLoRA: Federated Fine-Tuning Large Language Models with Heterogeneous Low-Rank Adaptations

Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild

SHED: Shapley-Based Automated Dataset Refinement for Instruction Fine-Tuning

Talks

February 02, 2025

February 02, 2025

February 02, 2025