Skip to content

Commit 79fde69

Browse files
johannag126johannag126
authored
Merge pull request #81 from IamShubhamGupto/master
new news and ran precommit on all files
2 parents 45abbfb + b55fad7 commit 79fde69

File tree

13 files changed

+67
-7
lines changed

13 files changed

+67
-7
lines changed

‎content/jobs/_index.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -18,7 +18,7 @@ Associate Research Scientist at the interface between climate modeling and machi
1818

1919
Postdoctoral researchers in Earth System Climate Modeling with AI at NYU. Available immediatly. [Apply here](https://apply.interfolio.com/140348)
2020

21-
### Princeton University/GFDL
21+
### Princeton University/GFDL
2222

2323
Postdoctoral Research Associate (or more senior scientist) using ML parametrizations for Ocean Data Assimilation Increments. [Apply here](https://puwebp.princeton.edu/AcadHire/apply/application.xhtml?listingId=32681)
2424

‎content/news/2312ClimSim.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22
date: 2023-12-12T09:29:16+10:00
33
title: 'ClimSim awarded Best Paper Award at NeurIPS'
44
heroHeading: ''
5-
heroSubHeading: ''
5+
heroSubHeading: 'ClimSim awarded Best Paper Award at NeurIPS'
66
heroBackground: ''
77
thumbnail: '/images/news/ClimSIm.jpeg'
88
images: ['/images/news/ClimSIm.jpeg']

‎content/news/2312WillGfeature.md‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -1,11 +1,11 @@
11
---
2-
date: 2023-12-11T09:29:16+10:00
3-
title: "Will Gregory's latest article featured in Advance Science News"
2+
date: 2023-12-01T09:29:16+10:00
3+
title: "Will Gregory\'s latest article featured in Advance Science News"
44
heroHeading: ''
5-
heroSubHeading: ''
5+
heroSubHeading: "Will Gregory\'s latest article featured in Advance Science News"
66
heroBackground: ''
7-
thumbnail: 'images/news/tina-rolf-yuF2B5Zyz88-unsplash.jpg'
8-
images: ['images/news/tina-rolf-yuF2B5Zyz88-unsplash.jpg']
7+
thumbnail: '/images/news/tina-rolf-yuF2B5Zyz88-unsplash.jpg'
8+
images: ['/images/news/tina-rolf-yuF2B5Zyz88-unsplash.jpg']
99
link: 'https://www.advancedsciencenews.com/ai-is-transforming-climate-forecasts-for-melting-sea-ice/'
1010
---
1111

‎content/news/2401L96.md‎

Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
---
2+
date: 2024-01-01T09:29:16+10:00
3+
title: "Learning Machine Learning with Lorenz-96"
4+
heroHeading: ''
5+
heroSubHeading: 'Learning Machine Learning with Lorenz-96'
6+
heroBackground: ''
7+
thumbnail: 'images/news/2401L96.png'
8+
images: ['images/news/2401L96.png']
9+
link: 'https://m2lines.github.io/L96_demo/intro.html'
10+
---
11+
12+
The M²LInES team is proud to share our Jupyter Book on [Learning Machine Learning with Lorenz-96](https://m2lines.github.io/L96_demo/intro.html). This educational tool provides a computationally accessible framework to understand how machine learning techniques can tackle climate science problems, including emulators, parameterizations, data assimilation, and uncertainty quantification. It is for all climate scientists wanting to learn or test machine learning algorithms or for machine learning experts to learn about climate modeling or develop new algorithms. The book was developed by and initially for our multidisciplinary M²LInES members, composed of machine learning experts, climate scientists, and numerical model developers. We hope you find it useful for all your research and educational needs!

‎content/news/2401Pavel.md‎

Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
---
2+
date: 2024-01-02T09:29:16+10:00
3+
title: "Data-driven equation discovery ocean model by Pavel"
4+
heroHeading: ''
5+
heroSubHeading: 'Data-driven equation discovery ocean model by Pavel'
6+
heroBackground: ''
7+
thumbnail: 'images/news/2401Pavel.png'
8+
images: ['images/news/2401Pavel.png']
9+
link: 'https://arxiv.org/abs/2311.02517'
10+
---
11+
12+
**Pavel Perezhogin** is lead author of this new M²LInES [preprint](https://arxiv.org/abs/2311.02517), describing the implementation, in the GFDL MOM6 ocean model, of a data-driven equation-discovery parameterization of mesoscale eddies. This scale-aware parametrization improved biases in the mean flow and energetics for a range of resolutions and in different ocean configurations. The work was done in collaboration with **Cheng Zhang**, **Alistair Adcroft**, **Carlos Fernandez-Granda** and **Laure Zanna**.

‎content/news/2402Bhouri.md‎

Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
---
2+
date: 2024-02-03T09:29:16+10:00
3+
title: "Stress-testing the coupled behavior of hybrid physics-machine simulations by Bhouri"
4+
heroHeading: ''
5+
heroSubHeading: 'Stress-testing the coupled behavior of hybrid physics-machine learning climate simulations on an unseen, warmer climate'
6+
heroBackground: ''
7+
thumbnail: 'images/news/2402Bhouri.png'
8+
images: ['images/news/2402Bhouri.png']
9+
link: 'https://arxiv.org/abs/2401.02098'
10+
---
11+
12+
Machine Learning (ML)-based parameterizations of atmospheric convection have long been hailed as a promising alternative, with the potential to yield higher accuracy at a fraction of the cost of more explicit simulations. This [work](https://arxiv.org/abs/2401.02098), co-authored by **Mohamed Aziz Bhouri**, investigates the coupled out-of-distribution extrapolation capabilities in "online" testing of different ML-based parameterization designs of atmospheric convection. Their results show that these design decisions are not enough to obtain satisfactory generalization benefits. The obtained improvement indicates the necessity of using multi-climate simulated training data.

‎content/news/2402Joan.md‎

Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
---
2+
date: 2024-02-02T09:29:16+10:00
3+
title: "Stochastic Optimal Control Matching by Joan"
4+
heroHeading: ''
5+
heroSubHeading: 'Stochastic Optimal Control Matching'
6+
heroBackground: ''
7+
thumbnail: 'images/news/2402Joan.png'
8+
images: ['images/news/2402Joan.png']
9+
link: 'https://arxiv.org/abs/2312.02027'
10+
---
11+
12+
**Joan Bruna** is a co-author in this [preprint](https://arxiv.org/abs/2312.02027), introducing Stochastic Optimal Control Matching (SOCM), a novel Iterative Diffusion Optimization (IDO) technique for stochastic optimal control. Stochastic optimal control’s goal is to drive the behavior of noisy systems. In this work, the control is learned via a least squares problem by trying to fit a matching vector field. The key idea underlying SOCM is the path-wise reparameterization trick, a novel technique that is of independent interest, e.g., for generative modeling. The code can be found on [github](https://github.com/facebookresearch/SOC-matching).

‎content/news/2402Noora.md‎

Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
---
2+
date: 2024-02-01T09:29:16+10:00
3+
title: "Restratification Effect of Mesoscale Eddies by Noora"
4+
heroHeading: ''
5+
heroSubHeading: 'Comparing Two Parameterizations for the Restratification Effect of Mesoscale Eddies in an Isopycnal Ocean Model'
6+
heroBackground: ''
7+
thumbnail: 'images/news/2402Noora.png'
8+
images: ['images/news/2402Noora.png']
9+
link: 'https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2022MS003518'
10+
---
11+
12+
**Nora Loose** is lead author of this [new paper](https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2022MS003518) - part of the Climate Process Team funded by NSF and NOAA - which compares two parameterizations (“GM90” and “GL90”) for the restratification effect of mesoscale eddies. The authors conclude that the less commonly used GL90 parameterization is a promising alternative to the popular GM90 scheme for isopycnal coordinate models, where it is more consistent with theory, computationally more efficient, easier to implement, and numerically more stable. **Alistair Adcroft** also contributed to the research.

‎static/images/news/2401L96.png‎

1020 KB
Loading

‎static/images/news/2401Pavel.png‎

1.98 MB
Loading

0 commit comments

Comments
 (0)