Recent years have witnessed the prevalence of large-scale data across different scientific disciplines and in various real-world applications. In particular, high-dimensional learning and inference have received growing amount of attention by many researchers in statistics and data science. Important fundamental questions include 1) how to pinpoint the precise theoretical properties for understanding the empirical success of popular learning methods such as deep learning, 2) how to understand the precise theoretical properties of popular learning methods such as random forests in general high-dimensional nonparametric models and design theoretically justified inference procedures, 3) how to deal with the common issue of missing data for minimax optimal classification, and 4) how to design double machine learning for multimodal data analysis. To address these fundamental challenges, this invited session brings together four experts who will introduce some cutting-edge developments on these interrelated topics from unique perspectives. A common theme of the session is high-dimensional machine learning and inference.

To address the aforementioned fundamental challenges, this invited session brings together four experts who will introduce some cutting-edge developments on these interrelated topics from unique perspectives. A common theme of the session is high-dimensional machine learning and inference.

Session organizer: Jinchi Lv , University of Southern California
Session chair: Jinchi Lv

List of session speakers (please arrange the talks in the order specified below):

1) Jianqing Fan , Princeton University
Talk title:  Factor Augmented Sparse Throughput Deep ReLU Neural Networks for High Dimensional Regression

2) Yingying Fan , University of Southern California
Talk title:  FACT: High-Dimensional Random Forests Inference

3) Timothy Cannings , University of Edinburgh
Talk title:  Minimax Optimal Classification under Missing Data

4) Xiaowu Dai , University of California, Los Angeles
Talk title:  Orthogonalized Kernel Debiased Machine Learning

Organiser: Dr Jinchi LV 

Chair: Mr Timothy Cannings 

Speaker: Dr Mahrad Sharifvaghefi 

Speaker: Mr Timothy Cannings 

Speaker: Xiaowo Dai