I Wrote a New Book!!!
Optimization: A Bootcamp for Machine Learning, Inverse Problems, and Control
Pre-Order Now (July 31)
https://t.co/EoDMFapUUf
Coming Soon:
* Free PDF on website
* YouTube Videos for entire book
* Python code on GitHub
Read this paper to know how to 'actually' read a paper!
I've highlighted the key points -> now this 10 min read of 3-pass apprach will change your paper reading technique for good!
๐๐ฒ๐ ๐๐ฎ๐ธ๐ฒ๐ฎ๐๐ฎ๐๐:
โ First Pass (5-10 minutes)
Quick scan for bird's-eye view
โ Second Pass (up to 1 hour)
Read with greater care
โ>> Third Pass below โฌ๏ธ
๐ฌ ๐๐ผ๐ ๐ฑ๐ผ ๐๐ผ๐ ๐ฎ๐ฝ๐ฝ๐ฟ๐ผ๐ฎ๐ฐ๐ต ๐ฟ๐ฒ๐ฎ๐ฑ๐ถ๐ป๐ด ๐ฝ๐ฎ๐ฝ๐ฒ๐ฟ๐?
1/2
ML weather models like GraphCast & NeuralGCM show promise in weather forecasting, but can they be used for data assimilation? Our study looks into their tangent linear & adjoint models, finding challenges in robustness & unphysical noise. ๐ฆ๏ธ #AI#NWP
https://t.co/0Hexm6dfSq
Finding relevant papers for literature review takes a lot of time.
So, MIT researchers built Undermind, an AI-powered search engine.
It can generate well-researched overviews on any topic and is 10-50 times better than Google Scholar.
Here's how to use it:
Many students have trouble understanding #Bayesian#MachineLearning, to help them I coded up Bayesian linear regression with MCMC to sample the posterior from scratch. Then I built out this interactive #Python dashboard with @matplotlib.
Vary the number of MCMC samples and watch the transition from burn-in to equilibrium chain. Try it out on #GitHub @ https://t.co/K4L8GjQwOU โ.
Limitation of Google Scholar: You can only look up articles with keywords. It doesn't work if you paste a sentence or a paragraph.
Sourcely is an AI-powered app that lets you look up papers using whole paragraphs.
Here's how to use it:
๐ซFellow scientific computing geeks, here is an example of adaptive mesh refinement in #JAX:
https://t.co/QOp1xW8I3S
I recently heard a claim that this was impossible in JAX due to its lack of dynamic shapes, and decided to prove the naysayers wrong ๐
Looking for a stats or applied maths PhD? There are 2 great PhD programmes at Imperial that are currently welcoming applications: "Mathematics for our Future Climate" and "Statistics and Machine Learning." Find out more here!...
https://t.co/GXjs37BAPc
https://t.co/dHzTAdzeLn
> Linear Algebra by Prof Gilbert Strang
> Probability Theory Prof John Tsitsiklis
> Information Theory by David Mackay
> Optimization theory by Stanford University
> Machine learning by Andrew NG(Stanford course)
This is all you need to master ML. Coursework + assignments + course projects and youโll be better than most people at ML.
As we're heading into PhD application season, I wrote a post with advice for applying to grad school, drawing from my experience on admissions committees.
Please share with any aspiring PhD students that you know! https://t.co/tthKWX8sNp
We just released our ETH Zรผrich AI in the Sciences and Engineering Masterโs course on YouTube! ๐
Prof. Siddhartha Mishra and I explain PINNs, neural operators, foundation models, neural PDEs, diffusion models, symbolic regression, and more! #AI4Science
https://t.co/bY26K9j16t