Education:

I received my Ph.D degree of “Climate Sciences” at School of Physics(page) of PekingUniversity supervised by Prof. Jun Wei(page). During my Ph.D career, I visited EAPS of MIT for one year, hosted by Prof. Paola Malanotte-Rizzoli(page).

My research spans hurricane, climate, physical ocean and AI. I focus on theoritically understanding the generalization of deep learning and applying ML to climate-ocean and hurricane (typhoon) prediction problems. I enjoyed making applications of AI and ML on climate, physical ocean, natural disasters, and so on.

I am working on a number of cross-projects in climate change, physic ocean, and CS. They usually involve coding in Python, Fortran and Pytorch. My detailed education and experience can be found in LinkedIn.

Research:

I’m interested in AI, climate science and physical ocean. My current focus is on deep learning theory and applying deep learning methods to the climate-ocean models.

7.Understanding Why Neural Networks Generalize Well Through GSNR of Parameters

Jinlong Liu, Guoqing Jiang, Yunzhi Bai, Ting Chen, Huayan Wang (2020)

Based on several approximations, we establish a quantitative relationship between model parameters’ GSNR and the generalization gap. This relationship indicates that larger GSNR during training process leads to better generalization performance. Moreover, different from that of shallow models (e.g. logistic regression, support vector machines), the gradient descent optimization dynamics of DNNs naturally produces large GSNR during training, which is probably the key to DNNs’ remarkable generalization ability.

Accepted as spotlight talk in ICLR 2020 in Addis Ababa, Ethiopia.

6.Seasonal and interannual variability of the subsurface velocity profile of the Indonesian Throughflow at Makassar Strait

Guo‐Qing Jiang, Jun Wei, Paola Malanotte‐Rizzoli, Mingting Li, Arnold L Gordon (2019)

The seasonal variability of the depth of the ITF velocity maximum is partially controlled by the seasonally reversed Karimata throughflow, while the remainder primarily originated from the Mindanao–Sulawesi inflow rather than the Sibutu Strait throughflow.

Accepted for publication in Journal of Geophysical Research: Oceans

5.Exploring the Importance of the Mindoro‐Sibutu Pathway to the Upper‐Layer Circulation of the South China Sea and the Indonesian Throughflow

Mingting Li, Jun Wei, Dongxiao Wang, Arnold L Gordon, Song Yang, Paola Malanotte‐Rizzoli, Guoqing Jiang (2019)

Closing the Sibutu Strait reduces the Luzon Strait throughflow into the SCS by 75%, and the Mindoro–Sibutu deep exchange is reversed, thus flowing into the SCS. No significant change occurs over the shallow Sunda Shelf of the southern SCS, which is primarily driven by local monsoon winds.

Accepted for publication in Journal of Geophysical Research: Oceans

4.Multi-decadal timeseries of the Indonesian throughflow

Mingting Li, Arnold L Gordon, Jun Wei, Laura K Gruenburg, Guoqing Jiang (2018)

NCEP reanalysis wind data from 1948 to 2016 and Makassar Strait transport from 2004 to 2011 are used to construct a multi-decadal timeseries of 0–300 m Makassar Throughflow using a back-propagation (BP) neural network.

Accepted for publication in Dynamics of Atmospheres and Oceans

3.A deep learning algorithm of neural network for the parameterization of typhoon‐ocean feedback in typhoon forecast models

Guo‐Qing Jiang, Jing Xu, Jun Wei (2018)

Two algorithms based on machine learning neural networks are proposed—the shallow learning (S‐L) and deep learning (D‐L) algorithms—that can potentially be used in atmosphere‐only typhoon forecast models to provide flow‐dependent typhoon‐induced sea surface temperature cooling (SSTC) for improving typhoon predictions.

Accepted for publication in Geophysical Research Letters

2.Parameterizing sea surface temperature cooling induced by tropical cyclones using a multivariate linear regression model

Jun Wei, Xin Liu, Guoqing Jiang (2018)

Combining a linear regression and a temperature budget formula, a multivariate regression model is proposed to parameterize and estimate sea surface temperature (SST) cooling induced by tropical cyclones (TCs).

Accepted for publication in Acta Oceanologica Sinica

1.Parameterization of typhoon-induced ocean cooling using temperature equation and machine learning algorithms: an example of typhoon Soulik (2013)

J Wei, GQ Jiang, X Liu (2017)

This study proposed three algorithms that can poten- tially be used to provide sea surface temperature (SST) condi- tions for typhoon prediction models.

Accepted for publication in Ocean Dynamics

---

Contact me: jianggq@pku.edu.cn