Scholar
Hao Cheng
Google Scholar ID: ftlVqVIAAAAJ
HKBU
Citations & Impact
Citations
2,027
H-index
16
i10-index
18
Publications
20
Co-authors
13
list available
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- Selected Publications:
- - RobustTSF: Towards Theory and Design of Robust Time Series Forecasting with Anomalies (ICLR 2024)
- - Unmasking and Improving Data Credibility: A Study with Datasets for Training Harmless Language Models (ICLR 2024)
- - Identifiability of Label Noise Transition Matrix (ICML 2023)
- - Mitigating Memorization of Noisy Labels via Regularization between Representations (ICLR 2023)
- - Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations (ICLR 2022)
- - Learning with Instance-Dependent Label Noise: A Sample Sieve Approach (ICLR 2021)
- - Pruning Filter in Filter (NeurlPS 2020)
- - Filter Grafting for Deep Neural Networks (CVPR 2020)
- - Local to Global Learning: Gradually Adding Classes for Training Deep Neural Networks (CVPR 2019)
- - Evaluating Capability of Deep Neural Networks for Image Classification via Information Plane (ECCV 2018)
Research Experience
- Spent two years at YouTu Lab, Tencent, working with Ke Li and Xing Sun;
- Research Intern at ByteDance AI Lab from 2023.03 to 2023.07 (working with Xiaoying Zhang, Yang Liu on Trustworthy Language Model);
- Research Intern at DAMO Academy, Alibaba from 2022.07 to 2023.02 (working with Qingsong Wen, Liang Sun on Robust Time Series Forecasting).
Education
- Ph.D. student at Hong Kong Baptist University (HKBU), advised by Professor Bo Han and Professor Yang Liu.
Background
- Research interests: enhancing the robustness of machine learning models, particularly by addressing challenges such as improving model resilience when trained on data with noisy labels. This encompasses a variety of data formats, including image, text, and time series data.
Miscellany
- Student organizer of IJCAI 2022 workshop on 1st Learning and Mining with Noisy Labels Challenge; Talk on IJCAI 2023 Tutorial: A Hands-on Tutorial for Learning with Noisy Labels
