Kaidi Cao

I am a Computer Science Ph.D. student at Stanford University, advised by Prof. Jure Leskovec. I did my bachelors at Tsinghua University.

My research interests lie generally in the area of Machine Learning, including graph representation learning and efficient, robust learning algorithms. Feel free to reach out to me through email (view page source if you are looking for sth).

Email / LinkedIn / Google Scholar / Github

Selected Publications
dise

Relational Multi-Task Learning: Modeling Relations between Data and Tasks
Kaidi Cao*, Jiaxuan You*, Jure Leskovec
International Conference on Learning Representations (ICLR), 2022
[pdf]

We propose MetaLink to solve a variety of multi-task learning settings, by constructing a knowledge graph over data points and tasks.

dise

Open-World Semi-Supervised Learning
Kaidi Cao*, Maria Brbić*, Jure Leskovec
International Conference on Learning Representations (ICLR), 2022
[pdf] [code]

We propose a pipeline that recognizes previously seen classes and discovers novel, never-before-seen classes at the same time.

dise

Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization
Kaidi Cao, Yining Chen, Junwei Lu, Nikos Arechiga, Adrien Gaidon, Tengyu Ma
International Conference on Learning Representations (ICLR), 2021
[pdf] [code]

We propose a data-dependent regularization technique for heteroskedastic and imbalanced datasets.

dise

Concept Learners for Few-Shot Learning
Kaidi Cao*, Maria Brbić*, Jure Leskovec
International Conference on Learning Representations (ICLR), 2021
[pdf] [code]

COMET learns generalizable representations along human-understandable concept dimensions.

dise

Coresets for Robust Training of Neural Networks against Noisy Labels
Baharan Mirzasoleiman, Kaidi Cao, Jure Leskovec
Neural Information Processing Systems (NeurIPS), 2020
[pdf] [code]

We propose a theoretically-principled method to create sets of clean data to train a model with noisy labels.

dise

Few-Shot Video Classification via Temporal Alignment
Kaidi Cao, Jingwei Ji*, Zhangjie Cao*, Chien-Yi Chang, Juan Carlos Niebles
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
[pdf] [split]

We propose a video few-shot learning framework that explicitly leverages the temporal ordering information in video data through temporal alignment.

dise

Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss
Kaidi Cao, Colin Wei, Adrien Gaidon, Nikos Arechiga, Tengyu Ma
Neural Information Processing Systems (NeurIPS), 2019
Oral presentation at the Bay Area Machine Learning Symposium
[pdf] [code]

We design two novel methods to improve imbalanced training.

dise

Learning Temporal Action Proposals with Fewer Labels
Jingwei Ji, Kaidi Cao, Juan Carlos Niebles
International Conference on Computer Vision (ICCV), 2019
[pdf]

We propose a semi-supervised learning algorithm for generating temporal action proposals.

dise

Delving Deep into Hybrid Annotations for 3D Human Recovery in the Wild
Yu Rong, Ziwei Liu, Cheng Li, Kaidi Cao, Chen Change Loy
International Conference on Computer Vision (ICCV), 2019
[pdf] [project page] [code]

We provided sufficient investigation of annotation design for in-the-wild 3D human reconstruction.

transgaga

Geometry-Aware Unsupervised Image-to-Image Translation
Wayne Wu, Kaidi Cao, Cheng Li, Chen Qian, Chen Change Loy
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019
[pdf] [project page]

We propose a geometry-aware framework for unsupervised image-to-image translation, which is robust to arbitrary shape variations between domains.

sigasia2018

CariGANs: Unpaired Photo-to-Caricature Translation
Kaidi Cao, Jing Liao, Lu Yuan
ACM Transactions on Graphics, (Proc. of Siggraph Asia), 2018
[pdf] [project page]

We present the first deep learning-based approach to automatically generate the facial caricature for a given portrait photo.

Press Coverage:
dream_mapping

Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
Kaidi Cao*, Yu Rong*, Cheng Li, Xiaoou Tang, Chen Change Loy
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018
[pdf][project page][code]

We presented a Deep Residual EquivAriant Mapping (DREAM) block to improve the performance of face recognition on profile faces.

cluster_pipeline

Merge or Not? Learning to Group Faces via Imitation Learning
Yue He*, Kaidi Cao*, Cheng Li, Chen Change Loy
AAAI Conference on Artificial Intelligence (AAAI, Spotlight), 2018
[pdf][code]

We proposed a novel face grouping framework that makes sequential merging decision based on short- and long-term rewards via inverse reinforcement learning.

Course Projects

Reinforcement Learning in Memory MAB
with Yujia Jin and Linjia Wu
[pdf]

Academic Services
  • Conference Reviewer: CVPR, ICCV, ECCV, AAAI, SIGGRAPH, NeurIPS, ICLR, ICML

  • Journal Reviewer: IJCV, TNNLS, PAMI

Teaching